Paper - Contribution to the study of the early human ovum

Embryology - 20 Apr 2024        Expand to Translate
Google Translate - select your language from the list shown below (this will open a new external page)
العربية | català | 中文 | 中國傳統的 | français | Deutsche | עִברִית | हिंदी | bahasa Indonesia | italiano | 日本語 | 한국어 | မြန်မာ | Pilipino | Polskie | português | ਪੰਜਾਬੀ ਦੇ | Română | русский | Español | Swahili | Svensk | ไทย | Türkçe | اردو | ייִדיש | Tiếng Việt    These external translations are automated and may not be accurate. (More? About Translations)

Johnston RW. Contribution to the Study of the Early Human Ovum based upon the investigation of I. A very early Ovum embedded in the Uterus and II. A very early Ovum embedded in the Infundibulum of the Tube. (1914) J. Obst. and Gynaecol. British Empire 26:

Online Editor
This historic 1914 paper by Johnston describes two early human embryos. Modern Notes:

Historic Disclaimer - information about historic embryology pages

Pages where the terms "Historic" (textbooks, papers, people, recommendations) appear on this site, and sections within pages where this disclaimer appears, indicate that the content and scientific understanding are specific to the time of publication. This means that while some scientific descriptions are still accurate, the terminology and interpretation of the developmental mechanisms reflect the understanding at the time of original publication and those of the preceding periods, these terms, interpretations and recommendations may not reflect our current scientific understanding.     (More? Embryology History | Historic Embryology Papers)

The Journal of

Obstetrics and Gynaecology of the British Empire

VOL. XXVI. MAY, 1 914. No.17!

Contribution to the Study of the Early Human Ovum based upon the investigation of

I. A very early Ovum embedded in the Uterus and II. A very early Ovum embedded in the Infundibulum of the Tube.

By

R. W. JOHNSTON JS, M.A., M.D., F.R.C.S.E.

Assistant to the Professor of Midwifery in the Uiiiversity of Edinburgh; Eater-n Assistant Physician, Royal M atemity Ilospital; Obstetric Physician, New Town Dispensary; Gyncecologist, Livingstone Dispensary.

OUR knowledge of the development of the human ovum during the first weeks of its life is still so incomplete that it becomes a duty to place on record the main features of any such specimen that may be met with. From the nature of things these specimens must be rare, and the majority of them will probably always -continue to be obtained in early abortions. Although the great possible value of such cases is evidenced by the flood of light that the Bryce-Teacher ovum has thrown on the subject, and by such a record as ]ung’s, still any specimen obtained in situ must of necessity approach more nearly to the normal. I consider myself peculiarly fortunate, therefore, in having obtained within ten months two specimens of an early ovum in situ/. Moreover the fact that one is embedded in the uterus and the other in the tube affords an interesting contrast and comparison of the earliest stages of development in the normal and abnormal situations.

The first specimen is that of an ovum of about fifteen days embedded in the uterus. Its general features do not differ in any

great degree from the classical description of Hubert Peters. The embryonic structures are, however, very difficult to interpret, and

16 232 Journal of Obstetrics and Gynaecology

one perfectly legitimate view of them is that the specimen is one of uniovular twins in a very early stage of development. If such be the case, the specimen is absolutely unique.

The second specimen is an infundibular tubal pregnancy of about twenty days duration. So far as I have been able to ascertain, this specimen is also unique, all the other records of the very early stages of tubal implantation being based upon ampullar or isthmic pregnancies. In addition to this the case has one other feature of great importance etiologically, namely that it affords a perfect demonstration of the fact that the ovum must have come from the opposite ovary and migrated round the back of the uterus to reach the site of its implantation.

I have thought it well to divide this paper into two parts. In the first place I have recorded the history and minute anatomical features of both specimens. In doing so I have tried to state what I saw, without reading into the statement my own opinion as to the nature of the appearances. The second part of the paper is a discussion of the different features, in which I have ventured to state my own views as to the specimens, and have compared their most interesting features with some of the other early ova on record.

TERMINOLOGY.

For the sake of brevity and to avoid misconception, it may be as well to state shortly the sense in which I shall employ c.ertain terms. The terms used to denote the topographical and histological divisions of the decidua, such as basalvis, capsularis, mm, compacta and spongiosa, require no explanation. The terms equatovrial. and polar applied to the decidua capisularis are explained in the text. Some confusion, however, exists in regard to the meaning of such words as syncytinm, trophoblast, trophoderm and the like.

The word brophoblast was applied by Hubrecht to the chorionic or extra—embryonic ectoderm. But strictly speaking it applies to it only when that structure remains in passive relationship to the surrounding tissues. VVhenever it proliferates and assumes histolytic functions it ceases to be in line with Hubrecht’s definition, and for this proliferative histolyti-c ‘trophoblast ’ Minot has suggested the term trophoderrm. In man, unlike many other mammalia, this layer is always actively proliferative, and accordingly the term trophoderm is the more correct one to use in human development. I agree with Herzog that etymologically either term is a misnomer. It suggests that the layer is wholly connected with the nutrition of the embryo. But as Herzog points out, the trophoderm is thousands of times greater in mass than the embryo, and this seems too extravagant a provision for nutrition alone. Indeed we have -clear evidence that nutrition is only a minor funcJ ohnstone: The Early Human Ovum 233

tion of this layer in the early stages, and its most important action is to enable the ovum to form stable attachments for itself to the maternal tissues. This it does by proliferating freely, and, in all probability, se-creting a proteolytic ferment which destroys and digests the maternal tissues with which it comes into contact.

The term syncytium has been so long applied to the outer layer of the trophoderrn as to be wedded to it. Etyrnologically it also is a misnomer, as the word connotes a tissue which is formed by the running together of cells. That this is the real mode of its formation We have no clear evidence; indeed there is as much reason to believe that it arises from the very beginning as a plasmodial layer. The term jnlasmvoditirophodevrm would be more correct on this latter understanding. Since, however, we have no proof on either side, and since the word syncytium is so constantly used in the literature, I have preferred to retain it. As will be explained in the text the term symplasma is used to denote a clearly degenerative coalescence of cells, or a degeneration of the syncy— tium.

FIRST SPECI)-lEl\'.

On March 16, 1913, a married woman, aged twenty—nine, died suddenly in this city. She had previously complained of no symptoms beyond a shortness of breath on exertion, and a tendency to faintness. She had had three children, the youngest a year old. The husband did not know the actual date of the last menstruation, but he was certain that she had not missed a menstrual period, and neither of them had had the slightest idea that she was pregnant.

A post mortem examination was ordered, and in the course of it the uterus was noticed to be somewhat enlarged. It was accordingly removed for further examination. The pathologist’s notes on the case are briefly as follows :~Tl1e heart was in a state of advanced fatty degeneration. The stomach was congested and empty. The liver, spleen, kidneys, and lungs were normal. The uterus was slightly enlarged, and soft. The tubes were engorged with blood. The right ovary was swollen and congested. The Cause of death was, therefore, the sudden failure of the heart, owing to advanced myocardial degeneration.

Unfortunately the autopsy was not made until nearly forty—eight hours after death, and the uterus was thereafter kept in Kaise1'1ing’s fluid for eighteen hours. It was then opened, but still other twelve hours elapsed before it was seen by Professor Harvey Littlejohn. He at once realised the possible significance of its contents, and sent it over to Dr. Barbour at the University Gynaecological Laboratory. In the absence on holiday of Dr. Barbour I began the investigation, and he has very kindly allowed me to continue and complete it. 234 Journal of Obstetrics and Gynmeology

The uterus was slightly enlarged, and the endometrium very distinctly thickened and thrown into irregular elevations and furrows. On the posterior wall of the cavity, a little above the centre, there was a small dark crimson object, protruding like a wart from the surface. It measured 46 mm. vertically, and 5 mm. from side to side. Its base was slightly narrowed, so that it overhung a little at its lower edge. In appearance it suggested a pedunculated wart, or a subepithelial hacmatoma, although the surrounding mucosa was not unduly congested. (1).*

A portion of the endometrium on the opposite side of the cavity was removed and embedded in paraffin. On section it showed the characteristic appearances of a decidua. The projection with its immediately surrounding decidua was thereupon removed, right down to the muscle wall of the uterus : and after I had obtained the advice and guidance of Professor Arthur Robinson on the specimen, it was dehydrated and embedded in paraffin. The entire block was then cut into 679 serial sections, each section being 10 microns thick. The plane of section was vertical to the surface of the decidua, and in the long axis of the uterus. The sections were stained with hsematoxylin and eosin.

The projection was found to contain a young ovum surrounded with early villi, lying in an implantation cavity partly filled with maternal blood, and bounded on all sides by decidua. (3 8: 4).

Examination of the sections revealed the fact that the specimen had unfortunately received considerable damage—~probably due to the uterus being firmly grasped antero—posteriorly during its removal from the body. There is thus a rent in the decidua capsularis, which has also extended into the wall of the blastocyst or chorionic vesicle. The latter -contains some maternal blood. showing that the damage had been received before the fixation of the specimen. Apart from this the specimen is in a wonderfully good state of preservation Considering the circumstances in which it was obtained. The histological appearances are not to any extent obscured.

DIMENSIONS or THE OVUM.

Owing to the fact that it is ruptured, it is im_possible to give absolutely exact measurements of the implantation cavity. But as the amount of displacement caused by the rupture is comparatively slight in most places, a closely approximate estimate may legitimately 19e made. In length from above downwards the implantation cavity measures (in section 21.1.1) 3'67 mm., and in depth from before backwards 2'46mm. It runs through 468 sect.ions, each 10 microns thick, and therefore measures in breadth 4168 mm.

Owing to the damage received by the specimen, much the same qualifications have to be made with regard to the measurements of the blastocyst as applied to the implantation cavity. The blasto—

• The numbers in brackets refer to the illiistrations.

FIG. 1.~—Uterus opened, showing ovum in situ-.

FIG. 2. —Corpus luteum. Johnstone: The Early Human Ovum 235

cyst, however, is not so much torn as the decidua capsularis, and accordingly a still more a-ccurate approximation to exactitude can be made in regard to it. Exclusive of the thickness of the wall, and of the outgrowing villi, the blastocyst runs through 272 sections, each 10 microns thick, and therefore measures 2,720 microns or 2'72 mm. from side to side. This is its greatest internal dimension. In section 22.1.5, whi-ch is the largest section of the blastocyst in which it is complete and untorn, the other two dimensions are respectively 1'66 and 084 mm. As this section is almost exactly through the middle of the blastocyst, it may be taken as giving the maximum dimensions with almost perfect accuracy.

On comparing it with the other youngest complete human ova, it is found to approximate fairly closely to that of Jung :—

Bryce-'l‘eacher, 077 x 063 x 0'52.

• Linzenmeier, 075 X 0615 X 0525.

Peters, 1'6 x 0'8 x 0'9.

Fetzer, 1'6 x 0'9.

v. Spee, 1'5 X 2'5.

Herzog, 232 x 0'8 x 1'2.

Heine—l-lofbauer, 2'38 x 0'98 x 1'2.

jung, 2'5 x 2'2 x 1'0.

johnstone, 272 x 166 x 0'34.

DECIDUA VERA.

The condition of the decidua vera was studied in the sections of a portion removed from the anterior wall of the cavity ————as far as possible from the site of the ovum. It measured 6'5 mm. in thickness. It presents in a characteristic way the appearance of two layers~—the superficial compact and the spongy. It is crammed with decidual cells, which are most easily observed in the compact layer, although present in the trabeculm between the glands of the

• Unfortunately, this )aper was in the press before Li11zenn1eier’s description of

Stoeckel’s ovum was published. The ovum was shown at the Gym. Congress at Munich in 1911. It was obtained, perfectly fresh, in srzibza in a uterus removed for repeated l1a‘,morrl1ages one month after curettage. The ovum is, therefore, less than a month old. The uterus was almost certainly pathological. In its stage of development the ovum would appear to lie between the Bryce-Teacher ovum and Peters’ ovum. The embryo consists of an amnion and yolk sac, the former measuring 0'l05><0'09 min., the latter half these figu1‘es. There is a definite embryonic disc measuring 0'21 x 0.105 x 0'02 mm. The embryonic anlagen are contained in a loose mass of mesoderm continuous with that lining the blastocyst, a distinct atlvance from the stage in the Bryce—Te-acher ovum. The mesoderm is just beginning to form cores to the villi. The intervillous space is filled with blood, fresh and apparently in recent circulation. The two layers of trophoderm are well defined, but towards the periphery of the trophodermic shell seem to merge into each other indiscriminately. The implantation cavity is closed by a closing coagulum, lenticular in shape, consisting of partially organised blood and fibrin and trophoderm cells. The endometrium shows no true decidual cells, but is in a state of premenstrlial congestion and oedema. The basal blood spaces are partly capillaries and partly blood-filled glands. The surface epithelium was intact except over the closing coagulum. The illustrations are disappointing. 236 Journal of Obstetrics and Gynaecology

spongy layer. No mitotic figures have been observed in the decidual cells, but this may well be due in part to the length of time that elapsed between tl1e death of the patient and the fixation of the specimen, as well as to the somewhat crude fixative used.

The surface epithelium is almost completely wanting. Here and there in the depressions around the mouths of glands a few columnar cells are to be seen in series, and here and there a few isolated cells are to be found sticking up aslant like the last remnants of a derelict palisade. Probably this is due to maceration for a similar condition prevails in the glands themselves. In them the epithelial cells are for the most part desquamated and lying loose in the 1umina—the whole decidua having a catarrhal appearance. Indeed the gland spaces are so much filled up with epithelial debris that the differentiation into the spongy and compact layers is slightly obscured at a first glance.

The blood vessels of the decidua vera are not unduly dilated. Here and there the veins are enlarged, but there is no great congestion, although the whole membrane is distinctly cedematous. Scattered throughout, but most marked in the proximity of the venules, there is a slight infiltration with round cells that look like

lymphocytes. DECID UA BASALIS.

The ovum is embedded in the superficial compact layer, and therefore lies upon a cushion of spongiosa and the deepest part of the compacta. This latter is, however, hardly recognizable as such owing to the presence of large blood spaces. The differentiation into the layers is not very obvious in the basalis, owing to these blood spaces, and to the catarrhal condition of the glands.

Decidual cells are to be seen in large clumps or islands in the compacta near the border zone. Elsewhere they are almost completely absent, and broadly speaking they are not so marked a feature of the basalis as of the decidua vera.

~ Immediately posterior to the ovum there are a great many large blood spaces crammed with maternal blood corpuscles. (4). In most cases these spaces may be recognized as distended capillaries or venules, either by tracing vessels into them and observing their sudden widening out, or by the recognition of the flat endothelial lining. In some cases where this recognition is impossible the spaces may be true interstitial spaces. In the border zone indeed the whole stroma is infiltrated with blood. In one or two cases the presen-ce of numerous degenerated epithelial cells in the blood suggests that the space was originally a gland space.

In the deeper portions of the decidua the congestion is less marked, and the normal corkscrew vessels are recognizable.

Throughout the entire decidua basalis, but most marked in the Fm. 3.—Section showing blastocyst and intervillous space.

B—Mesoderm of blastocyst shrunk away from DB——Decidua basalis.

epithelium in many places. ‘ DC-—Decidua capsularis. Bl—Bla.stocyst. i_ E—Embryonic vesicle with attachmenl to BS—- Blood space. : chorionic mesoderm.

CE-—Chorlonic epithelium. 5 IS—-Intervillous space.

Fla. 4. —Completc section of ovum and surrounding decidua, showing blood spaces opening into intervillous space.

BS~—Blood spaces in decidua ] M—BIa.stocyst wall (rup basalis. tm-ed). DB——Decidua basalis. O—~Blood _spaces opening

DC~—Decidua capsularis. directly into iutervillous IS—-Intervillous space. : space.

Fm. 6.——-At A in the decidua. cnpsularia gland tissue is present (see Fig. 6). A Johnstone: The Early Human Ovum 237

neighbourhood of the ovum, there is a definite leucocytic infiltration, in many cases considerable clumps of lymphocytes being seen in the proximity of a vessel.

The glands are very numerous and very dilated. The catarrhal condition of the epithelium has been referred to already. Towards the surface the glands appear to be pushed to the sides to make room for the blood spaces. They diverge and run up on either side into the capsularis, becoming more or less parallel to its surface.

DECIDUA CAPSULARIS.

One of the most noticeable features about the present specimen, and one which is obvious to the naked eye, is the marked extent to which the embedded ovum projects above the surface of the surrounding decidua. At the first glance it looked like a wart, and on closer examination like a pedunculated wart. This is borne out by the examination of the sections. The projection containing the ovum has a relatively narrow base, which it overhangs at the lower edge. (4).

It is convenient for purposes of description to follow Pfannenstiel in dividing the decidua capsularis into two divisions—the equato-rial and the polar. By the equatorial portions are meant those parts which form a ring around the ovum and are directly and immediately continuous with the decidua basalis. By the polar region is meant that part which covers the summit of the embedded ovum, and is farthest removed from the basalis.

In thickness the decidua capsularis gradually diminishes as we follow it from the equatorial to the polar region. Over the latter its average thickness is about 0'1 mm., while in the equatorial regions it attains as much as three times that thickness.

In the equatorial regions it consists of very much the same structures as the basalis—~that is to say, of oedematous endometrial strorna infiltrated with blood, and containing many distended vessels full of blood. As has alreadylbeen mentioned, the glands of the basalis are pushed aside by the wedge-shaped area of large distended blood spaces, and they run up into the capsularis for a short distance parallel to the surface. Disconnected portions of glandular tissue are, however, recognizable in the capsularis at a point very near the polar region. (5 & 6). I have not been able to recognize any gland opening into the implantation cavity.

The stroma of the capsularis does not contain any large decidual cells, but it is invaded particularly on its inner aspect and in the neighbourhood of the attachment of the cell-columns by tropho— dermic cells. The polar decidua consists very largely of fibrmoid material with a thin film of cells over its outer aSPt9Ct- TIPS fibr 1” is not altogether wanting on the inner aspect of the equatorial parts, 238 Journal of Obstetrics and Gynaecology

but it becomes distinctly more prominent over the summit of the ovum. It is infiltrated on its inner side by trophodermic cells, and in one or two places little masses of syncytium have penetrated almost through its entire thickness. One or two of these syncytial masses actually have the striated border, the nature of which will be discussed later.

THE BORDER ZONE.

As has just been indicated, the border zone on its capsular aspect—or in other words the inner edge of the capsularis~—~consists of endometrial stroma infiltrated with trophodermic cells. Here and there a layer of fibrin is noticeable, particularly over the polar region, but this is neither universal nor continuous. In many places the cells of the capsularis are very chaotically disarranged, as if the tissue had been torn up- by oedema, and many of them are more or less degenerated. There are numerous small spaces filled with blood, and in many cases these open directly into the implantation cavity.

The basal aspect of the implantation cavity is made up largely of the thin walls of the distended blood spaces. These walls are in some cases little thicker than films, composed of tissue which has undergone a fibrinoid degeneration and is invaded freely by trophodermic cells and plasmodium, In many areas this invasion has led to a loosening of the structure which has permitted the escape of blood into the implantation cavity. (22.1.3.) In short the implantation cavity is lined indiscriminately by invading trophodermic cells and plasmodium, by fibrinoid, and by more or less necrotic decidual stroma.

In at least two instances a distended blood vessel may be seen opening directly into the implantation cavity. (4). In one place the wall of a blood space appears to have degenerated. The cells have lost their outline and the nuclei are swollen, elongated and pale. The protoplasm contains brownish granules. Thus the tissue exactly «corresponds in appearance and staining reactions to the syncytium of the outer part of the intervillous space. (7).

THE CHORIONIC VILLI AND THE INTERVILLOUS SPACE.

All round the circumference of the chorionic vesicle villous processes are observed growing out towards the surface of the decidua. These villi are, practically speaking, equally numerous and equally long all round the chorion——~that is to say, there is not yet any increased exuberance of growth towards the decidua basalis, such as occurs later. The chorio-decidual space or intervillous space is full of maternal blood, which penetrates into all the interstices between the villi. F Ia. %.——Decidua. capsularis, showing remains of a gland. Below it is an elongated darkly stained mass of syncytium.

Fm. 7.—Seetion of small blood space in decidua basalis, showing degeneration of its well into "symplasma mate)-num conjunctivum." Fm. 8.——Portion of chorionic membrane with outgrowing villi.

C——Mesodermic core of villus T——Tropl1ode1-mic cell column. shrunk away from epi- W—~Spa.ce left by shrinkage of CH-«Che:-ion. [thelium. ‘ core.

Fm. 9.-—Showing attachment of trophodermic cell columns to decidua.

CC—Cell columns. V—Vil1us. DC-—Decidua capsularis. M—Meoderm of chorion.

Johnstone: The Early Human Ovum 239

The proximal portions of all the villi are hollowed out by a mesodermic core. In many places this core has, unfortunately, been withdrawn inwards from its epithelial covering by the shrinkage of the mesoderm during fixation. Where it has thus been separated from its covering it presents a very smoothly defined

_edge, not at all unlike an endothelium. There is no

appearance of embryonic capillaries in the villi. In addition to the -core the villi have a double covering of trophodermic epithelium-—— an inner layer of cytotrophoderm and an outer layer of plasmoditrophoderm. (8).

The cyt0trophoderm—Langhan’s laye1'—~is present mostly as a single layer over the surface of the chorionic membrane and the greater part of the villi. But towards the tips of the villi it becomes

‘ heaped together into large columns or masses. These unite with each other by irregular lateral outgrowths, and spread themselves

outwards to the surface of the implantation cavity in a shapeless and chaotic network. At the decidual surface they become attached by the trophoderm cells spreading laterally over the face of the decidua and also invading it. (9). The cells of the cytotrophoderm are

‘rounded or oval, and contain rather deeply staining round nuclei

which fill up the greater part of the cell.

The plasmoditrophoderm—the syncytium--——is very irregular in its distribution. In some p-laces it is absent over the surface of the chorionic membrane; in others it forms an almost unrecognizably thin layer; while in yet other parts it is very obvious as masses or buds, or as slender strands spreading over the surface of the cell columns, and stretching between them. (10). Farther out towards the decidua it is mostly found in such solid clumps or strands. On the surface of the chorionic membrane the syncytium where present has long, thin, flattened nuclei, diffusely stained. In the more uniform strands stretched over the cell columns and villi these oblong flat nuclei are still visible. But in the strands and clumps towards the periphery of the intervillous space the nuclei are more often larger, and rounded or oval in shape, vesicular, and faintly stained. Frequently these last nuclei are shrunken and crenated, and ghostly in appearance. The protoplasm of the syncytium is throughout foamlike and in many parts vacuolated. Everywhere this layer presents the appearance of tiny brownish granules, which are refractile, scattered through the protoplasm, and particularly numerous in the perinuclear protop-lasm. Sometimes these granules are so numerous as to give a strand of syncytium the appearance of a brown streak along the middle of it.

In the outer part of the intervillous space there are numerous large free cells, some with large pale nuclei, others with numerous small dark nuclei. The former can usually be traced by serial sections to the strands with the large degenerate nuclei referred to 240 Journal of Obstetrics and Gynaecology

above. The latter can also usually be traced to syncytial buds, the nuclei on cross section appearing smaller and round. Some of these wandering cells appear to be quite detached. In one or two instances the syncytium of these giant cells shows faint indications of a striated border.

There is a certain amount of vacuolation in the cell columns, empty spaces appearing, devoid of blood or any other content but a little structureless matter. These spaces are not due to the withdrawal of the mesodermic core from the villi, as they occur far beyond the utmost limits of the mesoderm in the substance of the cytotrophoderm columns. Serial sections do not show them to communicate with the intervillous ‘space. (11).

THE CHORIONIC MEMBRANE.

The epithelium has already been described in connection with the villi. Both epithelial layers stain deeply with haematoxylin and eosin, giving the ovum a dark outline even under low magnification.

The mesoderm of the chorion constitutes a fairly uniform layer of about 005 mm. in thickness. It consists of a deli-cate connective tissue with elongated spindle-shaped cells having round or oval nuclei arranged mostly parallel to the surface. The inner portion is less nucleated and gradually becomes continuous with a structureless debris inside the vesicle. This debris or ‘ magma ’ does not fill up the entire vesicle, but is quite irregular in its distribution. The outer edge of the mesoderm where it is shrunk away from the epithelium has a smooth appearance not unlike an endothelium.

EMBRYONIC ANLAGEN.

Owing to the damage received by the specimen the embryonic rudiments are considerably folded and at one or two points broken. It is therefore impossible to identify them with absolute certainty, and I shall in the first place merely describe what is to be seen.

As one follows through the sections one comes first upon a vesicle which is somewhat crumpled, but is roughly triangular in shape. (12 & 14). This vesicle passes through 35 sections and therefore‘ measures in this direction 035 mm. Its other maximum diameters are 089 and 0'21 mm. In most parts it presents two definite layers of cells. The outer layer, whi-ch is sometimes very much thinned, contains cells with oval nuclei. In the thinned parts these are flattened out and arranged parallel to the surface. This layer corresponds in every way to the mesodermic lining of the chorion, and with this tissue it is at one angle of the triangle definitely and substantially continuous. The inner layer is composed of rather larger cells which stain more deeply, and contain large round or oval nuclei with well marked, deeply staining chromatin network. FIG. 10. —'l‘r0phodermic cell columns, with strands of syncytium (syn.).

Fm. 11.—-Portions of tmphodermic cell columns with empty spaces (A) which do not co1'mnunicu.te with the intervillous space.

..-..._...... .. -. ,..... .._-........_....... . ....._. .._...._.........._.._........._..-...

FIG. 12.- «First vesicle of embryo.

CM-~-~-Clnoriollic lll(:S0d(').\‘ll1. '13-—-'I‘hic-keniug which may represent an embryonic area.

Fm. 18.——Both embryonic vesicles.

CM-—Mesoderm of chorion. | V,—Fi1-st vesicle. V2-—Second vesicle.

Johnstone: The Early Human Ovum 241

FIG. 14. First vesicle and its attachment to the mesoderm of the blastocyst wall.

FIG. I 5. First vesicle with pocket at lower right corner. B. Blastocyst wall. 242

Journal of Obstetrics and Gynaecology

FIG. 16.

FIG. 17.

Portion of blastocyst with both vesic1es.—Compare Fig. I 3,

Showing attachment of second vesicle to blastocyst wall. Johnstone: The Early Human Ovum 243

The shortest side of this triangular vesicle presents an appearance very like an embryonic area—that is to say, it is composed of cells heaped together in three or more irregular rows. In this area, which measures 0'21 by 0'2 mm., the nuclei are more oval, and tend to be arranged perpendicularly to the edge of the vesicle.

At one end of this ‘ area ’ a considerable fold or pocket is given off, which, but for serial sectioning, might easily have been mistaken for a second smaller vesicle. It is, however, directly continuous with the larger portion of the vesicle. (15).

FIG. 18. Portion of blastocyst with both vesicles, showing the flooring of cells in the second vesicle.

On the wall of the larger portion there are one or two thickenings which resemble the anlagen of blood vessels. But on tracing these through several sections they are found to be merely localised thickenings, which it is impossible to denote definitely as the precursors of vessels. This vesicle is unfortunately ruptured at one end, and there are a few maternal red blood corpluscles present within it, as Well as some of the mesodermic magma.

In the latersections containing this first vesicle a second vesicle makes its appearance to the left of the first. (13, 16,18). This passes through 26 sections and accordingly measures in this direction 0'26 mm. 244 Journal of Obstetrics and Gynaecology

Its other maximum diameters measure 0'28 and 0'14 mm. It is also somewhat -crumpled, and is roughly triangular in shape. In structure it is a replica of the first, presenting the same two layers of cells. At the angle farthest away from the first vesicle it also is definitely and substantially attached to the blastocyst wall, its outer layer being for some distance continuous with. the mesodermal lining of the chorion. (17).

In a number of the sections about one-third of the way through this vesicle, the cavity of the vesicle is almost wholly filled up with cells. (13 & 18). This flooring of cells appears to divide itmore or less completely into two smaller vesicles. This vesicle, like the first, is not absolutely intact, and several red blood corpuscles may be seen in its interior. At no point do the two vesicles come into actual contact, and their attachments to the wall of the blastocyst are towards the opposite extremes of the one side of that structure.

THE CORPUS LUTEUM. (2).

Exact measurements of the corpus luteum have not been made, but the figure shows a section through it as nearly as may be through the middle. This section measures 1'3 by 1 centimetre. The normal shape» is encroached upon by a small follicular cyst. The centre of the corpus luteum is occupied by a mass of pale finely granular material like mucus, stained pink with eosin. In it are several clumps and strings of red and white corpuscles. At the margins it becomes continuous with masses of red blood cells which fill up all the bays between the infoldings of the lutein layer. This layer is of varying depth but never less than eight or ten cells deep. It appears crenated owing to the papillary ingrowths, each of which has a stout ‘core ’ of young and extremely vascular connective tissue. These connective tissue ingrowths appear to be branching in parts and pushing the lutein layer before them. There is a great deal of vacuolation amongst the lutein cells, some considerably large spaces being formed. The cells are polygonal or star-shaped, and their terminal fibrils stretch across these spaces. The nuclei are about twice the size of a red blood corpuscle and stain well. The whole layer is very vascular, many capillaries being recognizable, and there is also a good deal of blood lying free in the stroma. At parts this blood is clearly continuous with that in the centre of the corpus luteum.

SECOND SPECIMEN.

The second specimen I owe to the kindness of Dr. William F ordyce. It was obtained from a private patient of his upon whom I assisted him to operate. She had a retroverted uterus, and an enlarged tender ovary on the left side. The abdomen was opened in order to perform a Gilliam’s operation, and to deal with the FIG. 19. ——Section through tubal ovum.

B»~B1astocyst (chorion). \ lS—Intervillous space, with BS»-Blood spaces infimbrias. I sections of villi." CM ~CapsuIar membrane. § T~Mass of trophodermic cells. F-—A-Fimbriae of tube. § YS—-Yolk sac (ruptured).

Fm. 20.—Capsular membrane. with trophodennic cells applied to the inner (right) surface.

Johnstone: The Early Human Ovum 245

ovary if necessary. The left ovary was found to be slightly cystic, and to our surprise there was a blood clot about the size of a large hazel nut protruding from the fimbriated end of the left tube. (30). The opposite ovary presented a recent corpus luteum on the surface.

The left tube and ovary were removed, and the round ligaments’

shortened.

There was nothing in the patient’s history to suggest, before the operation, the likelihood of a pregnancy, tubal or otherwise.

The tube and ovary were at once placed in Pick’s No. 1 solution and fixed. In the belief that the specimen was one of a tubal abortion, and that probably the ovum was broken up-, I opened the blood -clot after three days, with the intention of making a museum preparation. Somewhat to my surprise I found it to contain an apparently complete chorionic vesicle, although no trace of an embryo was to be seen by the naked eye. The vesicle seemed to be entirely surrounded by blood clot, but was entangled in or attached to the fimbriae anteriorly and below. I then cut away the infundibular end of the tube with the blood clot, and had serial sections made of the entire clot. The sections were cut 7 microns thick, and the Plane of section vertical and parallel to the long axis of the tube——this being the plane of my first incision. The sections thus pass from the back to the front of the specimen. In all some six hundred sections were cut. Most of them were stained with haematoxylin and eosin, but some with Mallory’s, some with van Gieson’s, and some with Weigert’s stains.

Study of the sections revealed an intact chorionic vesicle, containing an early embryo, superficially embedded in the fimbriae on the antero-inferior margin of the tube. The embryo is somewhat damaged, probably owing to the inevitable handling of the specimen during its removal, and partly to the unfortunate opening of the specimen before it was cut. (19).

The rest of the tube showed no traces, either macroscopic or microscopic, of having been the seat of a gestation, or of any other abnormality.

DIMENSIONS or THE OVUM.

The chorionic vesicle, exclusive of its villi, runs through about 390 sections. Owing to the fact that the specimen was opened with a razor before the sections were made, some small portion of the thickness appears to have been lost——not more than the thickness of one or two sections. But a close approximation would give its measure in this diameter as 2'73 mm. In one of the middle sections (52.1.1.) its other diameters measure 6 and 5'6 mm. respeetively. The whole vesicle is therefore of a flattened, almost circular shape. The flattening is in the diameter at right angles to the attachment to the fimbriae of the tube, and in this way it corres246 Journal of U bstetrics and Gynaecology

ponds to the lenticular shape of young ova embedded in the uterine decidua.

The embryo, including the two vesicles, is cut very obliquely, and runs through 306 sections. It therefore measures 2'lmm. in this direction. The embryonic area proper runs through 230 sections, and thus measures 1'6 mm.

ATTitC}IMENTS OF THE OVUM.

On the lower and anterior aspects the blood clot was seen to be attached to the fimbriae of the tube. Microscopic section shows that there is a definite organic attachment, thus indicating that the implantation was in all probability originally an infundibular one, and that we are not dealing with a tubal abortion, as at first I was led to suppose. The portions of the fimbrize which come into the sections are seen to be covered, on those aspects not in contact with the ovum, with intact columnar epithelium, ciliated in many parts. On the one side, however, the epithelium is absent, the ovum being definitely but superficially attached to it.

On this latter aspect there is a shallow saucer-shaped depression, to the surface of which the ovum has attached itself, and which represents, therefore, the implantation cavity. At the edges of this “basalis” area, if we may so call it, there has evidently been. an attempt at the formation of a ‘ capsular membrane,’ and some of the elements of the tube wall can be seen stretching a short distance round the ovum. Nowhere does the stroma of the fimbriw show any real decidual reaction in the shape of decidual cells, but there is in parts some enlargement of both muscle and connective tissue cells.

The whole stroma, moreover, is in a most chaotic state of disarrangement. It appears to have been torn up by oedema, the cells lying at some distance from each other, and many of the cells are distinctly degenerated. The blood vessels of the fimbriae are enormously distended and full of blood. Several of them form large blood spaces which are separated from the intervillous space only by a very thin wall. The stroma is in many parts densely infiltrated with red blood corpuscles.

The capsular membrane consists of stroma elements only for a very short distance on each side. For the most part these elements are strongly reinforced by fibrinoid material, which forms the sole covering over more than half the surface of the ovum. (20).

Bonner: ZONE.

In relation to the capsular membrane just described there is little to correspond to the border zone of a normal implantation beyond the presence of the fibrinoid material and the occasional Fm. 21. —Chorionic villus of twenty days‘ ovum.

Cap—--Capfllary being formed in mesoderm. I VT——Villus trunk. CC——Trophodermic cell columns. 1 CM—-Chm-ionic membrane. S-—SyncyL1'um. , F—Fimhria. V——Villus branch. \ IS-——Intervillous space.

J ohnstone: The Early Human Ovum 247'

invasion of it by trophodermic islands, either cellular or syncytialIn the ‘ basalis ’ area, however, in which there is maternal tissue present, the changes are closely similar to those in an intra-uterine implantation. Mention has already been made of the enlargement of some of the muscular and connective tissue cells, as well as of the dilatation of the blood vessels and the formation of blood spaces. In addition to this there is a very obvious invasion of the trophodermic elements into the tubal wall, both cells and syncytial masses being found at some little distance from the surface. There is also presumable evidence of the destructive action of the trophoderm in the presence of fibrinoid or semi-fibrinoid degenerated layers.

THE INTERVILLOUS SPACE AND THE CHORIONIC VESICLE.

The intervillous space is not full of blood except in some portions. But throughout there is clear evidence of its having contained blood in the presence of occasional blood islands, of red blood corpuscles adherent to the epithelium of the villi, and in the blood which fills many of the smallest interstices. The villi are distinctly more numerous, more luxuriant, and more than twice as long on the side of the chorionic vesicle apposed to the fimbrize of the tube. On the fimbrial aspect one villus measures 1‘75mm. in length, while on the opposite side of the blastocyst the average length does not exceed 0'75 mm. In other words there is here a commencement of the differentiation of the chorion frondosum and the chorion laeve.

In the chorion frondosum the villi are normal in their structure. I have been fortunate in getting one typical villus in longitudinal section through almost its whole length. (21). In this area also there is a great development and multiplication of trophodermic cells. These form large cell columns which branch outwards from the ends of the shorter villi, and, uniting with each other, form a chaotic network. Where they reach the surface of the tube wall or the capsular membrane they spread out laterally, and to a marked extent invade these structures. This appearance is much more marked on the fimbrial aspect of the ovum than on the capsular. On the latter many of the short villi end abruptly with no outgrowth of trophoderm. There are numerous spaces in these cell columns, some containing blood, others empty. Some of the empty spaces cannot be traced to the periphery of the columns, and are therefore enclosed spaces exactly similar to those described in the early intra-uterine ovum. Covering these cell masses there are strands and masses of plasmoditrophoderm or syncytium, to which reference will be made later.

The blastocyst consists of a layer of mesoderm covered by two layers of epithelium. All three layers are continued into the villi.

17 248 Journal of Obstetrics and Gynaecology

The inner layer of trophoderm epithelium is Langhans’ layer, and consists of a very regular single row of cells with rounded or oval nuclei which fill up the greater part of the cells and stain deeply. It is noticeable that where these cells pass into the cell columns at the ends of the villi, as just described, they become distinctly enlarged.

In the -chorionic membrane and the villi the syncytium is spread as a thin covering of fairly uniform thickness over the Langhans’ layer. The protoplasm has a finely foamy appearance. The nuclei are mostly small and oval with a well marked chromatin network, and deeply stained nucleoli. A few of the nuclei are larger and pale, but I have not observed any so ghostly in appearance as many seen in the earlier intra-uterine specimen. In some of the more tangential sections Langhans’ cells appear in two or three layers, as also the nuclei of the syncytium. I believe, however, that this is largely due to the obliquity of the sections.

In the intervillous space syncytium is present, as mentioned, in the form of strands and masses on the outside, and not infrequently in the inside, of the trophoderm cell columns. In several places there are islets of multinucleated syncytium lying free. Some of these may be traced by serial sections to buds of the villous coverings; others cannot, and appear to be absolutely free.

An interesting point about the syncytium in this specimen is that in many portions——in practically all the buds and islands just described and in quite a number of the villi——it presents a distinct cuticle which has a striated appearance.

Inside the trophoderm the blastocyst is lined with a layer of delicate mesoderm, varying in thickness. This passes out into the villi forming their cores. Strangely enough the structure of this mesoderm is more definite in the villus cores than in the portion lining the chorionic membrane. In one or two of the cores capillary vessels may be seen in the process of formation. In none of them was any blood observed.

The structure of the mesoderm is similar to that of the earlier specimen. The nuclei are more or less oval, and towards the outside are arranged parallel to the surface. Internally it fades irregularly into a structureless magma which is present here and ‘there within the vesicle. At the corner of the vesicle corresponding to the greatest development of the villi, the mesoderm reticulum is rather greater in quantity, and here is to be found the embryo.

THE EMBRYO.

The yolk sac has been ruptured, and part of its wall displaced outward. Including this displaced portion the whole embryo runs through 306 sections and therefore measures 2142 mm. Exclusive of this, the embryonic structures are present in 230 sections, which \.~.§_-i/ Fm. 2‘2.—~Embryo-—hind sections.

A——Amni0tic cavity. | C—Choi-ion. BS——Bc-lly stalk. 5 G~—Gui. M ——]\/ledullary groove.

F19. 2:'5.—Allantois (All.) just visible, passing from hind gut into the belly stalk.

\\»-‘_ __.,.’—¥"/ / FIG. 24 A ~-- Amniotic cavity. ' C —-Chorion. All -«Allantois. G-— Gut.

M——Medullary groove.

C—CoeIom. | M—«—Medu1lury groove.

FIG. 26.

A‘—~Amnion. M——Medullary groove. Coe1.—-Coelom. YS—Yolk sac, ruptured and G—-Gut. opened out.

H-—Hea.rt.

Fm. 27.-—Syncytia.l mass, showing the striated border.

J ohnstone: The Early Human Ovum 249

corresponds to 1'61 mm.—a more accurate measurement than the former. It has been -cut very obliquely in a dorso-ventral plane. The hind portion is well seen, but more anteriorly the sections are not so satisfactory. In great part this is due to imperfect technique in mounting the sections, but the result has been to obscure some interesting appearances, and to render it difficult to be certain as to the absolute normality of the embryo. All that can be said is that those parts which can be well seen are normal.

In the earlier sections we find the caudal end of the embryo. The medullary groove, the notochord, the mesodermic somites, and the hind gut are all clearly visible. (22). Closely surrounding the whole is the amnion, consisting of two layers of cells. Posteriorly the hind gut dips down to the surface, from which it is separated by an anal plate or membrane. Below the embryo the amnion is attached to a mass of mesoderm~the belly stalk. This contains one large patent vessel with definite walls, and one or two small capillaries. The large vessel contains a number of red blood corpuscles, the great majority of which are definitely nucleated. In a small minority the nuclei have not taken on the stain satisfactorily. More anteriorly the gut becomes widened out, and the dip down to the anal plate disappears. (23 & 24). The belly stalk appears in direct continuity with the mesoderm of the body wall, the folds of the amnion coming down to the sides of the stalk. [n the belly stalk the allantois may be distinguished as a slender column of epithelial cells running from the gut into the stalk. A little more anteriorly the gut is seen directly continuous with the widely opened yolk sac. The coelom is just beginning to appear. The medullary groove is closed. (25).

At this point we come upon some unsatisfactory sections, but in more anterior sections we see the gut separated off from the yolk sac which is opened out widely and ruptured. (26). On each side the coelom is visible, and projecting into the two halves are the anlagen of the heart. The amnion at this level is closely applied to the surface of the embryo. The yolk sa.c consists of two layers, much thinned out. At parts there are thickenings which seem to be the anlagen of vessels. The more anterior sections are, unfortunately, damaged.

Comparing these appearances with the data in the Noirmentafeln it is legitimate to conclude that the embryo comes in between the

“ Gle ” embryo of v. Spee and the “ Klb ” embryo of Kroemer and Pfannenstiel.

DISCUSSION. The Decidua.

The main features of the so-called ‘ decidual reaction’ are the transformation of the embryonic connective tissue corpvuscles of 250 Journal of Obstetrics and Gynaecology

the stroma—-this origin can no longer be reasonably doubted—~into the large decidual cells; the increase and distension of the glands, which become more tortuous in their course; the increased thickness of the endometrium, which becomes thrown into folds on the surface; and occasionally the congestion of the vessels and the oedematous nature of the stroma. VVhat the stimulus is which gives rise to this profound -change is unknown, as also the time of its occurrence—whether before or after the implantation of the ovum. Much speculation has been devoted to the question whether the stimulus may be due in some obscure bio-chemical fashion to the act of fertilisation, or to the influence of the corpus luteum, or to the actual invasion of the mucosa by the developing ovum. On none of these points have we any certain knowledge, but there are one or two facts from which probable conclusions may be drawn.

It is well known, for example, that analogous changes on a less marked scale take place in the endometrium during the premenstrual phase. It is also probable that a considerable number, if not the majority, of fertilised ova are embedded during that stage. Bryce and Teacher have argued strongly that the ovum may be embedded during any phase of the menstrual cycle, and this is in all likelihood the case. At the same time the premenstrual stage would appeal‘ to be the one in which most ova with a definite history were implanted.

ln the second place it is noteworthy that in several of the recorded specimens of early ova the decidual reaction has been very imperfect. Thus the differentiation into the compact and spongy layers, which depends upon the increased growth of the glands, is present in Peters’ case only in the immediate neighbourhood of the ovum, and in the older Heine—Hofbauer specimen it is wanting altogether. Again in both Peters‘ and ]ung’s ova there was no clear formation of decidual cells, but only a few enlarged cells in the vicinity of the ovum.

Furthermore we know that in the mouse, and more particularly in the guinea-pig, the endometrium undergoes no change whatever until the ovum has actually eaten its way through the surface epithelium.

Putting these facts together it seems not unfair to conclude that in all probability the decidual reaction proper does not occur until the ovum begins to invade the endometrium, and that the slight changes present in some recorded cases are merely the _persistence of premenstrual conditions of varying intensity. The rapidity with which the reaction ensues may quite naturally vary in different cases.

It. must not be forgotten that a strong argument against this provisional conclusion appears to rest in the formation of a uterinedecidua in cases of extra-uterine pregnancy. But there is noJohnstone: The Early Human Ovum 25]

evidence, so far as I am aware, to indicate that the uterine change in such cases occurs before the ovum has embedded itself in its abnormal situation, and therefore we may legitimately conclude that the change is again due to some bio—chemical influence consequent upon the invasion of the ovum.

This argument practically assumes the truth of the view, first promulgated by Sir VVilliam Turner,that the decidual reaction is a protective one on the part of the maternal tissues. Everything points to the correctness of this view. There is, for example, the effort at decidual formation in the tube itself during pregnancy in that situation, as well as the appearance of decidual cells in the ovary, peritoneum, and cervix in some non—pregnant conditions. In all cases, moreover, the appearance of decidual cells is accompanied by a leucocytic infiltration, as has recently been pointed out by Meyer, and therefore it may be regarded as in some measure comparable to the ordinary reactions of inflammation. If the biochemical stimulus be blood—borne from the trophoderm, as seems most plausible, the uterine reaction in tubal pregnancy may easily be explained as a praiseworthy but misplaced effort on the part of the maternal tissues to counteract it.

At the same time the decidual reaction is in all probability indirectly but equally protective to the embryo, in that, by supporting and strengthening the maternal capillaries, it prevents a too sudden and too extensive opening up of these vessels—the probable consequence of which would be to tear up the delicate attachments of the ovum. The formation of a decidua is therefore to the mutual advantage of both mother and child.

My first specimen shows a very characteristic decidual reaction. The differentiation into the compacta and the spongiosa is well marked, although it is somewhat obscured by the free desquama— tion of the epithelial cells lining the glands. Decidual cells are found throughout in the compact layer, and to a lesser extent in the trabeculae of the spongy part. I failed to find any mitotic figures in the cells, but this may be due to the circumstances in which the specimen was obtained. As a rule such changes are only to be seen in perfectly fresh and carefully fixed preparations. Marcliand, Bonnet and others have recorded the presence of many. On the other hand, Herzog, whose specimen was obtained and prepared in an almost ideal manner, failed to detect any karyokinetic changes.

The comparative absence of the superficial epithelium in the present specimen is, I believe, largely due to maceration. It is absent in the Bryce-Teacher specimen, as might be expected after its adventures. It is wanting also to a great extent in C0va’s ovum in spite of good preservation. In the great ma]or1ty of other cases,

however, it is present and in fairly good condition. 252 Journal of Obstetrics and Gynaecology

This maceration, which is evident also in the deeper parts of the glands in my ovum, makes it impossible to recognize the papillary projections of the epithelium into the lumina of the glands, which are by some regarded as a characteristic feature of the early decidua.

Apart from this maceration the glands seem to have suffered but little change even in the decidua basalis. They have obviously been pushed aside by the expansion of the ovum, the wedge—shaped space between them being occupied by the “cushion ” of dilated blood spaces. Bryce and Teacher suggest that the glands are more resistant to the destructive action of the trophoderm than the other tissues, and this may well be so, although they ultimately stiffer some dissolution. Frassi has shown conclusively that the glands may be opened into laterally, and it is probable that this may be effected either by the eroding action of trophodermic buds, or by the digestive action of the proteolytic ferments of the tropho— derm, while that tissue itself is at some little distance.

This is presumably the explanation of three observed facts——« Firstly, the presence of epithelial “ rests,” in the “border zone," formerly but probably erroneously regarded as relics of the superficial epithelium; secondly, the opening of glands into the intervillous space as observed by Frassi and Fetzer~the glands as they run up towards the equatorial portion of the decidua capsularis being opened into laterally; and thirdly, as a consequence of this the presence of blood in the gland spaces, which has been noted by Siegenbeek van Heukelom, Leopold, and Fetzer amongst others.

These conditions I have not observed in the present specimen. The blood spaces nowhere have, unequivocally, the characters of glands, but for the most part appear to be dilated vessels. The flat endothelial lining can be seen in many places, while in others the identity can be established by tracing fully formed vessels into the spaces, and noticing how these vessels suddenly widen out into sinuses. Indeed they are absolutely comparable to the blood spaces in the second (tubal) ovum, where there is no question as to the possibility of their being glands.

In at least two places the blood spaces can be seen opening into the implantation cavity or intervillous space. At these points there is no indication of any proliferation of‘ the endothelium over the surface of the implantation cavity. Nor is there any definite or continuous lining of that space by syncytium. On the contrary the space is lined somewhat indiscriminately by fibrinoid tissue and trophodermic elements.

The oedematous condition of the decidual stroma is noted in most early ova, but it is usually associated with a certain amount of congestion of the blood vessels throughout the whole decidua. Jung refers particularly to a condition of venous hyperaemia. In Johnstone: The Early Human Ovum 253

my specimen there is no marked congestion in the decidua vera, but merely some enlarged veins here and there. There is in the decidua basalis, especially around the border zone, a noticeable escape of red blood corpuscles into the stroma. This condition is more widespread in my specimen than in ]ung’s, where it was confined to the immediate neighbourhood of the border zone. I have not detected anything that could be interpreted as the formation of new capillaries in this zone. This phenomenon is described by Peters, particularly in the capsularis. White blood cells, or at least what appear to be lymphocytes, are to be found in the stroma particularly around the vessels, not only in the decidua basalis but in the vera and capsularis.

The embedding of the uterine ovum.

In connection with the decidua capsularis I have already mentioned that the embedded ovum projects to a quite unusual degree above the surface of the surrounding decidua. The only other specimens which resemble it in this respect are those of Leopold, Cova, Marchand and Rossi Doria—particularly the lastnamed. This circumstance has led Cova to suppose that the ovum was necessarily arrested by a projecting piece of the endometrium, upon which it proceeded to engraft itself. Rossi Doria, on the other hand, finds in it support for the older View of the process of embedding, namely that the ovum comes to rest on the surface of the mucous membrane and is gradually enclosed by an upgrowth of decidua capsularis around it. He modifies it, however, and states his belief that the implantation is effected by a combined method——the ovum eroding its way through the superficial epithelium into the substance of the mucous membrane, while at the same time the surface decidua grows up round it as a true ‘capsularis.’

In the present specimen the projection of the ovum is even more marked than in Rossi Doria’s specimen, and, as has been said, gives it almost the appearance of being pedunculated. At the same time I do not consider that there is any need to explain this by any peculiarity in the mode of implantation. Cova’s View can neither be substantiated nor denied. It remains a mere pious opinion. On the other hand there is no evidence to support Rossi Doria’s view. There is, it is true, no mushroom-like stopper of fibrinoid material (“gewebspilz”) at the point of entrance, as in Peters’ and some other cases; but there is clear evidence that the polar portion of the decidua capsularis consists practically entirely of fibrinoid material~the natural interpretation of which is that it represents the stretched out “closing coagulum ” at the point of entrance of the ovum. Moreover there is absolutely nothing to indicate any active growth of the decidual elements in the decidua 254 Journal of Obstetrics and Gynaecology

capsularis, such as one would expect if it were a-ctively growing up around the ovum. There is, for example, no sign of cell division.

I believe itito be more likely that there was originally a closing coagulum at the point of entrance, but that it has disappeared partly by absorption and partly by stretching. Furthermore the projection of the ovum is probably to be explained by two things~ firstly, that the ovum may have been rather superficially embedded, and secondly, that as a result, the direction of least resistance, in which growth would naturally take place, was, even more markedly than usual, towards the surface.

Another interesting point in the present specimen, which at the same time goes to prove that the capsularis is no new formation, is the presence of relics of glandular structure in it well up towards the polar region. This is clearly shown in the photographs. (5 & 6).

Embedding of the tubal ovum.

At this stage it is interesting to compare and contrast with the conditions in the early uterine pregnancy those of the second specimen—the early tubal pregnancy.

The trend of a good deal of the recent work upon tubal gestation has been to throw the responsibility for this abnormality more upon the ovum than upon the tube. The old idea of preceding disease of the tube is only rarely borne out by an examination of the affected tube. In a great many cases the tube is remarkably healthy. This is so in the present instance, the rest of the tube showing no abnormality, nor trace of disease.

VVhen we look to the ovum for an explanation, we are forced to conclude that in some circumstances it must be fertilized at or near the ovary, and that its development goes on so rapidly, or it is so delayed in reaching the uterus, that it becomes provided with the necessary trophodermic apparatus for embedding itself while still in the tube. In the present instance there is a particularly beautiful illustration of the truth of this view. In the ovary removed along with the pregnant tube there is no recent corpus luteum. On the ovary of the other side there was observed at the time of operation what we took without hesitation to be a recent corpus luteum. The ovum must therefore have travelled from the one side of the uterus to the other. Further, assuming that it was fertilized on the surface of the ovary from which it arose, and that the journey to the mouth of the opposite tube occupied several days, it is possible that by the time it arrived at this point it was ready to implant itself. Under these circumstances it is less surprising that it engrafted itself on the fimbriae, instead of passing along to the ampullary portion of the tube, which is so much more common a site for implantation.

It must be admitted that the assumptions thus made are in the present state of our knowledge not susceptible of proof. We have J ohnstone: The Early Human Ovum 255

no actual knowledge of the frequency or rarity of the presence of spermatozoa in the peritoneal cavity, although the possibility of it is proved by cases of ovarian and primary abdominal pregnancy. Again we have no proof that the journey from the one side of the uterus to the other must occupy the ovum for several days. Lastly we have no certain knowledge of what, in the case of man, brings the ovum to rest either in the tube or the uterus, or of the condition of the developing ovum prior to the blastocyst stage. As a matter of fact these considerations are rather lost sight of in all theories of tubal gestation which seek to place the whole blame upon the tube.

It is legitimate to conclude that when the human ovum escapes from the ruptured Graafian follicle it is surrounded by the zona pellucida, and outside that by some of the immediately adjacent cells of the discus pmligems, the so-called corona radiata. Through these two coverings the spermatozoon penetrates, and development begins. The subsequent fate, as well as the function, of the corona radiata is unknown. In his Hunterian Lectures Professor Robinson has suggested that its cells may provide pabulum for the growing ovum while the latter is still in the higher reaches of the Fallopian tube. On the other hand it may go to form an albuminous coating, such as is found in the ova of several animals—~the dog, and the rabbit, for example. This layer by its stickiness probably acts as a means of fixing the ovum to the mucous membrane. VVe do not know, however, that any such layer is formed in the human ovum, and apart altogether from it we are still in the dark as to how long the zona pellucida persists. In some anima1s~Tupaja javanica according to Hubrecht~it dis~ appears at the two—celled stage. In the mouse and rat it persists to about the eight-celled stage, and even later in the bat. In the mole and the shrew it persists until the embryonic ectoderm comes to the surface of the ovum, while in the dog and the ferret it does not disappear until after the appearance of the primitive streak and the -commencement of the formation of the mesoderm. Generally speaking it disappears early in those animals in which the ovum is early embedded in the decidua.

The functions of the zona are somewhat obscure. In all probability, according to Robinson, it is not so much a protection from actual pressure as from premature attachment. Thus in animals in which the embryonic ectoderm comes to lie on the surface of the ovum the zona persists until the embryonic ectoderm cells have reached a stage of development at which they no longer tend to fuse with the decidua. On the other hand, in animals such as the mouse, rat, guinea-pig, hedgehog, and bat, in which the embryonic ectoderm never comes to the surface but remains separated from the maternal tissues by other cells or a fluid-filled space of considerable size, the zona disappears at an early stage. 256 Journal of Obstetrics and Gynaecology

In these animals its function is probably just to prevent the union of the chorionic cells with the decidua until the ovum has attained a sufficient size. To this second class there is reason to believe that man and the apes also belong.

Bearing in mind the possibilities lying behind our ignorance of these points it becomes apparent that in this case. and in a great many other cases of tubal pregnancy, the arrest of the ovum in the tube may be due to the premature or the excessive development of the hypothetical albuminous layer around the ovum, or to the premature disappearance of the zona pellucida. Of the conditions governing any of these circumstances we are totally ignorant.

But to return to the present instance, I believe that my assump— tions, although they cannot be proved, afford the simplest explanation of the conditions found——namely that the journey of the ovum from the right ovary to the mouth of the left tube~—a journey considerably longer than that to the interior of the uterus by the ordinary route~occupied the ovum for several days, and that by the time it reached the tubal infundibulum it was ready, in a perfectly normal sense, to embed itself. Moreover if this occurred in the case of a woman with tubes that were, so far as could be seen, perfectly healthy, it is probable that a similar explanation underlies quite a number of other cases of tubal pregnancy.

The actual process of embedding in the case of tubal pregnancy is now regarded as being as strictly analogous to what occurs in the uterus, as the differing circumstances will allow. There is every reason to believe that the ovum eats its way through the superficial epithelium of the tubal mucosa exactly as it erodes its way into the endometrium. The first difference met with is the absence of anything like a thick stroma underneath the epithelium of the tubal mucosa. In most cases therefore the ovum eats its way right into the muscular wall of the tube, and forms there its implantation cavity. This is probably a slower process than sinking into the substance of the soft endometrium, and accordingly the surface epithelium does not get the same chance of closing over it as the decidua capsularis has in the uterus. Thus the tissue which separates the ovum from the lumen of the tube degenerates almost as soon as it is formed, and is present in most specimens as a membrane composed largely of fibrin. To it is applied the name “capsular membrane." In some cases, however, the ovum sinks deeper, and several authors have demonstrated tubal muscle elements in the capsular membrane.

In many ca.ses there are obvious attempts at the formation of a decidua. The relative absence of a connective tissue stroma makes it impossible for this to be a marked feature in the tube, but even in the presence of such stroma the reaction is often absent, and if present is only in small isolated patches. The vascular changes in Johnstone: The Early Human Ovum 257

the immediate neighbourhood of the ovum, and the effects of the action of the trophoderm are analogous to what is found in the case of uterine pregnancy.

The above applies in the main to ampullar pregnancies, which form the majority of tubal gestations. But it will be seen that, mutatis mutandis, it explains all the appearances of the present specimen.

Thus the saucer-shaped hollow in the fimbrize, in which the ovum rests and to which it is attached, shows that the ovum had eroded its way through the Surface epithelium into the substance of the fimbrize. This portion of the fimbria or fimbria: represents the decidua basalis, and is the seat of a marked reaction similar to what is found in the true decidua basalis. There are no decidual cells, but merely an enlargement of some of the muscle and stroma elements. The whole tissue is markedly oedematous, and there is a striking enlargement of the blood vessels with the formation of very large blood spaces separated from the intervillous space only by thin films of tissue.

A capsular membrane has been formed, which in the ‘equatorial’ parts presents the structure of the fimbriae. Over more than half the ovum, however, this structure is replaced by blood clot woven through with coarse streaks of fibrin, and irregularly interspersed with Cells. Most of the latter appear to be trophodermic in nature. VVhether this capsular membrane originally consisted to a greater extent of fimbria] tissue which has subsequently undergone fibrinoid degeneration, or whether it has been from the outset largely made up of blood clot, it is impossible to say. The appearances incline me to the latter view, and I think the large expanse of fibrinous clot must be regarded as corresponding to a closing coagulum at the point of original entrance of the ovum, which has been excessively stretched.

This in itself, however, is not a sufficient explanation, as it cannot be held to include the whole of the blood clot that was seen at the time of operation, and is shown in the photograph. To cause this there must have been some bleeding, and although I have not been able to detect the source of the bleeding in the course of the microscopic examination, yet I think it must have come in some way from the surface of the equatorial portion of the capsular

membrane. It is likely that the blood oozed out very slowly, and

having percolated gradually round the ovum formed the clot. In doing so it probably reinforced the earlier fibrinoid capsule just referred to. Some such occurrence as this is at least possible considering the very shallow implantation of the ovum, and the fact that, as has been shown in my other specimen as well as by many other observers, the vessels of the decidua may be opened up by the influence of the chorionic elements at some little distance from the actual chorionic cells. 258 Journal of Obstetrics and Gynwcology

In this connection I gave full consideration to the possibility of the specimen being, as I thought probable at the first, one of a tubal abortion—that is to say, of an ovum which had been implanted farther along the tube, and had then been dislodged and expelled to the abdominal ostium without the embryo being killed. and which had then engrafted itself secondarily upon the fimbriae. As already stated, however, the rest of the tube appeared perfectly healthy, and in the parts examined microscopically there was not the faintest sign of any undue congestion, still less of any destruction of the lining epithelium. The normal outline and shape of the tube were unaltered. Further, the clot, on the outside at any rate, appeared quite fresh and recent, and therefore must have been formed long after the ovum had attached itself to the fimbriae. It is, of course, generally agreed that after such an abortion the tube very quickly returns to its normal condition, but I hardly think it could do so in quite such a short time as might have been available in this case, or to such a complete extent in that time. I believe therefore that I am right in regarding this as a primary infundibular pregnancy.

From the nature of the site of implantation there can be nothing to correspond to a decidua vera, and, as has been stated, there is practically no change in the rest of the tube. Even the vascular engorgement is very localised and does not extend beyond the affected fimbriee. In this it resembles the intra-uterine specimen. where the engorgement was practically confined to the decidua basalis.

The surface epithelium present on the distal side of the fimbriae shows no sign of enlargement or of degeneration or maceration. Even the cilia can be readily detected.

I have not detected any blood vessels opening directly into the intervillous space.

The Border Zone.

The border zone or transition zone or “Um1agerungszone" marks the division between the foetal and maternal tissues. As might be expected it is not a sharp line, for it is in truth the fighting line where the conflict between the maternal. cells and the invading trophoderm takes place, and it is strewn with such of the dead on both sides as have not already been carried off the field or otherwise disposed of. In the earlier stages there is even more evidence of the deadly nature of the conflict than in the later stages. The Bryce-Teacher ovum, for example, has a very marked zone containing degenerated and necrotic cells of both maternal and foetal origin. But in the later stages there is evidence that the maternal tissues are succeeding in limiting the invasion. Thus we do not Johnstone: The Early Human Ovurn 259

find the large degenerated “maternal” cells described by Bryce and Teacher in the margin of the intervillous space.

Put briefly, the border zone in the first specimen consists of decidual stroma very much torn up by oedema or by the influence of the trophodermic elements, and invaded every here and there by these elements. This invasion is most noticeable where the cell columns of trophoderm are attached to the surface of the cavity. There is thus an irregular distribution of trophodermic cells all round the cavity, but no definite layer of syncytium such as is found later——the “basal ectoderm” of Langhans. Streaks and patches of fibrinoid material are also irregularly present all round the cavity, and no doubt represent Nitabuch’s fibrin layer. This tissue is, however, not sufficiently regular in its distribution to warrant one in regarding it as a third layer separating material and foetal elements, as is claimed by Fetzer.

Jung and F rassi have pointed out that the infiltration of the decidua with leucocytes enables one to differentiate between the foetal and maternal tissues, because these cells are unable to penetrate the foetal area. In my specimen there is a general infiltration by leucocytes through the decidua, and particularly in the neighbourhood of the ovum. At the same time their presence is not sufficiently marked to enable me to regard them as a satisfactory boundary line.

I have mentioned that the decidual tissue is in parts so opened out as to permit the escape of blood from the underlying blood spaces. I have also mentioned that the whole tissue is infiltrated with red blood corpuscles in the neighbourhood of the ovum. The result is that round the edge of the implantation cavity there are numerous tiny bays or pools of blood into which the red cells seem to be filtering through the decidual stroma. The edge of the cavity thus somewhat resembles the sea shore at low tide, when there are many small pools communicating with the sea on the one side and receiving innumerable small runnels or streams of water from the Other side.

There is also a clear connection on a larger scale between the blood in the implantation cavity and the maternal blood spaces. In two instances there can be seen opening into the -cavity blood vessels, which can be traced in the other direction into the large blood spaces of the decidua basalis. In this respect the specimen agrees with those of Frassi and Cova.

The tubal specimen agrees in the main with what has been said of the uterine one. The streaks of fibrin are on the whole more uniformly present. This is particularly so on the inside of the capsular membrane. At the same time trophodermic cells may be seen far beyond it, and it can in no sense be regarded as a dividing line between foetal and maternal tissues. 260 Journal of Obstetrics and Gynzeeology

A very interesting point of difference is that in the tubal specimen the invasion of trophodermic elements into the tube wall is distinctly more marked. Detached masses of both syncytium and Langhans’ cells may be seen at quite a considerable distance from the edge of the gestation cavity, and, furthermore,the natural arrangement of the maternal tissues in the border zone is much more destroyed. The whole tissue in the zone is indeed ploughed up to a quite extraordinary extent.

It seems quite fair to assume that this is due to the absence of the decidual cells, and of the support and resistance which they lend to the maternal tissues. At any rate this is a feature that has been noted frequently in tubal gestations, where deportation of portions of syncytium along the blood stream is not an unusual occurrence.

The stroma of the tube wall in this zone is infiltrated with red blood cells, but there is not the same formation of small bays opening into the intervillous space, nor have I seen any vessels opening directly into the space. The leucocyte infiltration is a much less marked feature of this specimen.

The I-ntervillous Space .

In the uterine specimen the intervillous space is quite full of maternal blood. In the tubal specimen it is not so, but the space contains several considerable blood islands, as well as innumerable individual red -corpuscles, clinging to the surface of the villi, and lurking in many of the smaller interstices. This forms clear evidence of its having been full of blood.

There can be little doubt that this is the normal condition, and that there must be a slow circulation by means of such channels as have been described. The force of the blood stream must be broken by the intervention of the large blood spaces in the decidua basalis, and the danger of a forcible escape of blood into the intervillous space during these days of slender attachments of the ovum greatly obviated. The arrangement is indeed an ideal one for securing a safe and slow circulation,

Several specimens have been described in which there was no blood in the intervillous space, and on the strength of such a specimen, otherwise apparently normal, Marchand actually stated the opinion that it was the normal condition. The presence of blood in the Peters’ ovum, for example, he regarded as due to the effects of the caustic alkali poisoning from which the patient died. Marcl1and’s own specimen, however, was obtained from a woman who died of severe internal haemorrhage, and this may explain the absence of blood from the intervillous space. Since then Herzog’s specimen, obtained under almost ideal conditions, has been found to have the space full of blood; while F rassi, whose ovum did not Johnstonez The Early Human Ovum 261

contain blood, regards that as an abnormal feature, due in all probability to the squeezing of the uterus during its removal by vaginal hysterectomy.

The fact that the blood, although outside the vessels, does not clot, has been pointed out again and again. It is the only instance in the body of such a phenomenon in connection with normal (non-menstrual) blood. The explanation would seem of necessity to lie in some attribute of the trophoderm elements—both the cellular and the plasmoidal layers—which prevents clotting. Hofbauer has suggested that this may be of the nature of a fine layer of albumose secreted by the trophoderm.

The further contents of the implantation cavity are in both cases the villi, and the outgrowths of the trophodermic epithelium. In both specimens the epithelial covering of the chorionic membrane is very similar—— a single layer of Langharfs cells regularly arranged, and outside that a thin layer of syncytiuin. In the tubal specimen, which is obviously older, there is a beginning differentiation into chorion laeve and chorion frondosum. In one or two parts also the villi of the second specimen show the commencement of capillaries in the mesodermic core.

In the first specimen all the villi, and in the second case the shorter ones which do not reach to the surface of the tube wall, -terminate in great outgrowths of trophoderm. These cell columns branch very irregularly, uniting with each other laterally, and reaching out to the surface of the implantation cavity to which they attach themselves. At the points of attachment the cells invade the maternal tissues as has been described.

Tm: “THIRD LAvi~:R.”

Before passing on to the discussion of the trophoderm, a brief reference must be made to the question of the existence of the much debated third layer between the two-layered trophoderm and the decidua. In some of the lower animals a series of giant cells has been described lying between the trophoderm proper and the maternal tissues. They first appear about the time that the embryonic mesoderm is formed. In the mouse and the hedgehog, for example, they are frankly phagocytic, and portions of red blood corpuscles and decidual cells may be seen in their interior. Hubrecht accordingly called them deciduof-mots, and they are regarded as a means of enlarging the implantation cavity,

In the mouse, to quote from Lochhead’s admirable summary, “ Duval stated that each was derived from a cell of the foetal ectodermal wall of the yolk sac, and later from a cell of the ectoplacental cone. As they increase in number, they form a distinct layer between the yolk sac and the wall of the implantation cavity, and some wander into the decidua and lie singly or in groups. ln their 262 Journal of Obstetrics and Gynaecology

interior degenerating leucocytes are frequently seen. Sobotta also stated that they were foetal in origin, and helped to fix the ovum and erode the maternal capillaries. More recently Kolster has brought forward evidence from their histological appearance that they are transformed decidual cells, and this is strongly supported by Disse’s investigations on the field-mouse, in which the giant cells are found before the ovum has become embedded, and the first to appear are at an appreciable distance beneath the surface epithelium. A second series of smaller size appears later in the lumen and wall of the implantation -cavity. jenkinson also recognized two groups, but assigned to them different origins, foetal in the “ Eikammer ” and maternal in the decidua.

All authorities agree that they are phagocytic. The tissue around them undergoes fatty degeneration, and in their interior may be seen remnants of connective tissue and endothelial cells and fat globules. Many capillaries are ruptured, and red and white blood corpuscles are also absorbed. Such an absorption of maternal tissue by the giant cells leads to an increase in the size of the implantation cavity, and a thinning of its wall. In spite of their abundant supply of nutriment, their life-history is short. No cell divisions occur, and they soon degenerate. Their contents are absorbed by the trophoblast, and their protoplasm shrinks to form a rim around the nucleus. Later still their remnants are also absorbed.”

I have quoted this passage at length to show how dubious are the nature and origin of these cells even in lower animals, in which such a problem is much easier of solution than in man. Bryce and Teacher have described a number of somewhat similar cells in their human ovum, “lying free in the blood space within the necrotic layer of the decidua ” in small excavations or bays. “ Within the layer itself are seen spaces enclosing cells in different phases of degeneration whose nuclei are identical with those of the completely free cells, and the protoplasm stains the same dusky.red in both cases.” They admit that “these cells closely resemble bodies which are certainly cross-sections of plasmodial strands, but when traced through the sections it becomes quite certain that they are not continuous with the plasmodiuin, but are really isolated cells.” Largely because they agree in nuclear characters and protoplasmic reactions with cells farther out in the decidua which are obviously degenerating maternal cells, the authors -come to the conclusion that they are more likely maternal in origin. The cells are, they state, “ quite different from the wandering trophoblast cells of later phases.”

In Peters’ ovum there are also some large cells which the author compares to the deciduofracts of the hedgehog. Siegenbcek Van Heukelom, too, describes such a layer, and regards its cells as Johnstone: The Early Human Ovum 263

foetal, derivatives from the cytotrophoderm. “ VVhatever their origin,” says Lochhead, “ the mononuclear cells in man appear to be engaged in disintegrating mucosal tissue, and producing a zone of coagulation necrosis, i.e., a symplasma, around the trophoblast. But they differ from similarly situated cells in lower animals, e.g., the mouse, in showing no evidence of ingestion of formed tissueelements.”

The conclusion of the whole matter would therefore appear to be that it is a perfectly open question whether the cells described in the early human ova are really comparable to the deciduofracts of the lower animals. My own specimens throw no light whatever on the problem. There are, as has been described, numerous wandering giant cells in the outer part of the trophodermic sheath and intervillous space, but these are all so closely similar to the cross—sections of plasmoditrophoderrn strands, that their origin from these cannot be held in doubt, even although they are quite detached and isolated. It is, however, quite possible that such cells do occur in the earlier stages, such as the Bryce-Teacher ovum, and very soon thereafter disappear. As has been stated, they are very short-lived even in the lower animals.

The Tmphoderm.

The cytotrophoderm cells (Langhans’ layer) correspond closely in both ova. In the chorionic membrane they are arranged in a single row. This arrangement is continued over the trunks of the villi, but at the tips of such villi or their branches as do not reach the periphery of the intervillous space, they become heaped up into the cell columns described. Where the arrangement is regular the cells are rounded, and the round or oval nuclei occupy most of the cell space; but in the cell columns they are so massed together that the shape becomes in many cases irregularly polygonal. The size of the cells also is greater in the columns, and where they spread out over the surface of the maternal tissues they have a superficial resemblance to decidual cells. Their direct continuity with Langhans’ layer as well as a closer study of their appearance, however, make it quite clear that, as has also been shown by Williams, Berkeley and Bonney, and a host of other observers, they are not maternal in origin.

In both specimens these cells invade the maternal tissues, and may be detected at some distance from the edge of the implantation cavity. This is particularly the case in the second specimen, and is probably, as I have indicated, due to the absence of the decidual reaction.

A comparison of the syncytium, or more correctly the syncytiotrophoderm, in the two specimens reveals several very interesting points.

1.8 264 Journal of Obstetrics and Gynaecology

In the uterine gestation it is spread so thinly over parts of the chorionic membrane as to be hardly recognizable. On the outside of the villi it is more uniform in arrangement, and amongst the cell columns it is to be found in clumps and strands. (10). The protoplasrn is always foam-like or finely granular, and more or less studded with fine brownish. granules. In only one or two instances is there any suggestion of a striated border. Over the chorionic membrane and villi the nuclei are mostly long, flattened oblongs, staining diffusely. Elsewhere this type of nucleus is replaced in parts by large, pale, and often -crenated and shrunken nuclei, Containing one or two darkly staining nucleoli. Numerous apparently detached islands may be seen, some containing the large pale nuclei, others groups of the small, dark, round ones, probably the oval ones on section. In many cases these islands may be traced through serial sections to the strands mentioned. Others are quite detached, but there can be no reasonable doubt that their origin was the same.

In the tubal gestation the syncytium is more uniformly arranged over the outside of the entire chorionic membrane, and is always easily recognizable. On the outside of the villi it is more exuberant in growth, and the intervillous space contains a much greater number of buds and detached islands. As before some of the islands or giant cells can be traced by serial section to villus buds, while others are entirely detached. The protoplasm is again very definitely foam-like, and stains uniformly with eosin, and apparently also with the haematoxylin, to judge from the purplish-red hue. The nuclei are preponderatingly rather small and dark and are very numerous. A well marked striated border is visible in many places. Over the chorionic membrane it is but seldom seen; over the villi it is more frequent; whilst it is best and most constantly seen on the syncytial buds and the detached giant cells. Nowhere is there any appearance of the brown granules so uniformly present in the other specimen.

Of all tissues in the whole realm of histology the syncytium has almost certainly been the most discussed during the last twenty-five years. Peters, whose classic description of his ovum, threw the first definite light upon it, mentions some ten distinct theories as to its nature and origin I Since that publication it has been demonstrated with a gradually increasing clearness that the human ovum is embedded in much the same way as in the case of the guinea-pig, that is to say, interstitially in the uterine mucosa. This knowledge at once threw out of -court at least two of the more prominent views of its origin—namely from the surface epithelium of the endometrium, and from the glandular epithelium. Afurther and still more conclusive proof of the erroneousness of these views has been obtained more recently in the numerous carefully described Johnstone: The Early Human Ovum 265

cases of ovarian pregnancy, in which the syncytium is found to be identically the same as in an ordinary intra—uterine pregnancy.

This leaves only two views of importance, one that it arises from the endothelium of the vessels of the decidua, and the other that it is foetal in origin. The former view was last championed by Pfannenstiel, all other observers having abandoned it in favour of the latter. Recently, indeed, Frassi has given it the coup de grace by showing from Pfannenstiel’s own figures that what that observer regarded as endothelium may with much greater probability be looked on as an epithelial rest~that is, a portion of epithelium remaining in a gland, which has been opened into laterally by the trophoderm, and converted into a blood space. This occurrence has been referred to already in connection with the nature of the blood spaces in the decidua basalis.

In one area my first ovum shows an appearance which might readily have been taken to support Pfannenstiel’s view—namely a blood space lined with “ syncytium.” (7). The cells have lost their outlines and the nuclei are elongated and pale. The protoplasm contains the brownish granules which are found throughout the syncytium in this specimen. I believe, however, that this tissue is not syncytium in the sense of plasmoditrophoderm, but rather a degeneration of the maternal tissues lining the space. In other Words, it corresponds to what, as I mention below, Jung describes as “ symplasma maternum conjunctivum.”

The foetal origin of the syncytium is indicated in a much more positive way in such an ovum as Bryce and Teacher’s. In it there is no connection between the syncytium and the maternal tissues, whereas it is in intimate connection with the blastocyst wall.

W'hile, then, it is generally agreed nowadays that the syncytium is the outer layer of the trophoderm, we have no exact knowledge of the earlier stages of its development or of its essential nature. Peters regarded it as the reaction of the cytotrophoderm to Contact with the maternal blood, and both he and Jung have described appearances which led them to think that it is formed out of the cellular layer underlying it. The only appearance in my specimens which might be taken as supporting this view, is the presence of syncytial islands in the centre of some of the cell columns. This appearance can, however, also be explained by the fusion of two parallel cell columns, and the consequent enclosure of the syncytium between them.

Hubrecht and Bonnet regard the syncytium as an indication of intense functional activity on the part of the trophoderm when it comes into -contact with the maternal structures. This implies an organic reaction of a more or less automatic kind between the foetal and maternal tissues, just as Peters’ theory does. VVhether or no this be the -case, the term “ Contact ” is certainly too restrictive. 266 Journal of Obstetrics and Gynaecology

For there are large stretches of syncytium that never come into contact with the maternal tissues—such for example as the layer covering the chorionic membrane in between the villi. It is true that in some cases, as in my first specimen, the syncytium is irregular in its distribution over that area, but still it is present to some extent, and in many other specimens it forms a complete covering. Even in the earliest 0vum—the Bryce—Teacher specin1en~there is much syncytium that is not, and probably never was, in contact with the maternal tissues. Accordingly the stimulus to the reaction, if such it be, must be more far-reaching than actual contact.

From the study of my own specimens I am not inclined to regard the syncytium in the sense of a reaction of the trophoderm cells to contact or even association with the maternal tissues or blood. Such a reaction would be universal. But we find cytotrophodermic elements in intimate Contact with the maternal tissues and blood without their being changed into a syncytium. On the other hand the syn-cytium has everywhere, except in the degenerate forms, the appearance of being an intensely active tissue, designed primarily for the dissolution of the maternal tissues, and possibly later for the less aggressive functions of selecting and transmitting nutriment. I would therefore subscribe to the view that the syncytium is probably the outer layer of the trophoderm from the very outset. This does not, of course, exclude the possibility of its being formed continuously from the cytotrophoderm. Such an origin is most probable, but it is not specifically called out by association with the maternal tissues. Apart from this question of its origin, Grosser’s view as to its changing functions, which will be mentioned in a later paragraph, seems the one which best suits our rather conflicting knowledge of the tissue.

Before elaborating this latter view, however, it is necessary to consider and contrast. the syncytium as present in such an early specimen as the Bryce-Teacher ovum and as present in older ova.

The Bryce-Teacher ovum is surrounded by a loose spongework of syncytiotrophoderm, which “forms an extraordinarily extensive spun-out investment for the ovum. It. occurs in masses, bands, or threads. The nuclei are invariably small and stain darkly.” Peters’ ovum represents a stage subsequent to this. In it the syncytium is distributed around the blastocyst in an extremely irregular fashion, but the “spongework " has largely disappeared. In the still older ova the arrangement becomes increasingly uniform as a layer on the outside of the cytotrophoderm, and the only trace of the previous arrangement is to be found in the irregular strands and masses interspersed among the cell -columns towards the periphery of the intervillous space.

Some of these latter masses are regarded by Jung as degenerated Johnstone: The Early Human Ovum 267

portions of the syncytium or of maternal tissues as the case may be. These he terms symplasmata~-~either symplasma syncytiale foetale or symplasma maternum conjunctivum according to their apparent origin. Another point of possible distinction between the earlier and the later syncytium is that the latter shows a striated border in many cases.

It will be seen that 1ny first specimen corresponds to _Iung’s description. The masses and strands of syncytium containing large shrunken nuclei are what he calls symplasma syncytiale fcetale, and the fact that in my specimen these nuclei are pale and ghost-like, while in his they stain diffusely, is probably due to a more advanced degeneration. One instance of the symplasma maternum I have already referred to. (7). In the ;~‘.ec0nd specimen there is very little of this symplasma formation. On the other hand the striated border is well marked in it—both possibly points indicative of a more advanced stage of development.

Grosser makes what seems a helpful and thoughtful suggestion when he speaks of there being “so to speak ” two generations of syncytium. “The first is associated with the implantation and with the intense histolysis which leads to the formation of the cavity of the egg capsule; it is represented by the syncytial shell of the Bryce—Teacher ovum.” As time goes on this early syncytium may be supposed to degenerate into the symplasma syncytiale, which has all the appearance of a degeneration product. Meantime the histolytic function is largely usurped by the cytotroph0— derm, which is seen invading the maternal structures to a much greater extent than the syncytium. The newer generation of syncytium, on the other hand, may be considered to assume either a secretory or an absorptive function, of which the striated border is probably an indication. This supposition involves the constant formation of new syncytium, and this can most easily be accounted for by the view that it is formed continuously from the cytotropl1oderm. The appearances noted by Peters, Leopold, and Jung suggesting an actual stage of transmutation confirm this. At the same time it is probable that the syncytium has in itself powers of reproduction, for syncytial buds may be seen in the placenta long after the cells of Langhans’ layer have ceased to exist. Still, the disappearance of the cellular layer does largely coincide with the time when the demand for the formation of syncytium in large quantity begins to diminish, and accordingly the cellular layer may well be regarded as to a great extent, although not exclusively, the mother-layer of the syncytium.

In this connection I should like to point out that, even assuming the truth of this view, there may be other reasons associated with the relatively early disappearance of Langhans’ layer in the placental villi. Lochhead has indicated one possible theory, which 268 Journal of Obstetrics and Gynaecology

is suggestive. He points out that during the early weeks of development, in which both coverings are present over the villi, the different organs of the foetus, indeed the different tissues, are hardly sufliciently developed to be able to conduct their own metabolic activities, even on a limited scale. More selective discretion must therefore be exercised for them——in other words there must be a more rigid filtering and selection of the substances passed through the villus epithelium to the foetal blood. ln the later months, on the other hand, the organs become increasingly more and more able to select their own particular forms of nourishment, and to Carry on their own individual metabolism. The efficiency of the placental filter may therefore be relaxed. This stage coincides with the gradual disappearance of Langhans’ layer.

The striated border of the syncytium (27 &28) has been studied mainly by Bonnet, Marchand, and Lenhossek. It has been described as a row of prickle processes—the Biirstenbesatz—and as a ciliated margin. Lenhossek studied it in a fresh specimen and maintains that the processes are not motile, but that they are provided with basal granules in the cuticle, which may be observed in specimens stained with iron haematoxylin. No one else has described such granules, and I have entirely failed to see them even in specimens stained with iron hzematoxylin. The striated margin has, however, been described in quite a number of the young ova, including those of Jung and I-‘Ierzog. Peters is not definite on the point but speaks of “ a delicate and thin, strongly refractive deposit, slightly frayed at the edges, on the surface of the syncytium.” Rossi Doria regards it merely as a deposit of broken-up blood corpuscles on the surface of the syncytium.

Peters’ description applies closely to many areas in my tubal ovum, and in regard to Rossi Doria’s View it must be admitted that there is frequently a deposit which is quite recognizably made up of broken-up blood cells. But in many other parts the appearance is more definite, and, moreover, I have seen it in syncytium buried in the substance of the decidua -capsularis, where it would be impossible to imagine an accidental deposit of broken-down blood cells.

The tubal specimen affords the opportunity of contrasting in one and the same section this striated cuticle and the ciliated surface of the tubal mucosa, which is still quite obvious on the outer side of the fimbriae. The comparison shows that the tubal cilia are more slender, as might be expected since they are motile; whereas the “ rods ” of the striated border appear stiffer and straighter, and do not suggest the possibility of movement. In a word, they are more rod—1ike than the cilia themselves.

The question arises whether in the absence of any power of movement these processes are really rods, or whether the appearFm. 28.—St1-iated border of syncytium. On the lower edge several adherent blood corpuscles may be seen.

Fm. 29.—Syncytial strand surrounded by maternal blood. The foam-like structure of the syucytium may be seen, also the dark granules scattered through it. The outline of the syncytial strand is lost, as the lens was focussed on the

granules. Johnstone: The Early Human Ovum 269

ance may not be due to the presence of pores out of which secretion is passing. Against their being merely pores is their extraordinary regularity as well as the straight stiff appearance mentioned. At the same time it is difficult to conceive of their being unassociated with some function either of secretion or absorption, and accordingly it is quite possible that the spaces in between them represent pores.

Apart from such functional reason for their existence it might be conceivable to regard them as a phylogenetic memory corresponding to the ciliated epithelial covering of some elementary marine forms of life. But this is harking back to a very remote ancestry indeed, and against it must be placed the fact that, so far as our knowledge goes, this appearance is not found in the earliest known syncytium (as in the Bryce-Teacher ovum) but only in the somewhat later stages.

The appearance of small refractive brownish granules scattered throughout the protoplasm, and more particularly the perinuclear protoplasm, of the syncytium in my first specimen is a little dillicult to explain (29). Exactly similar appearances are figured by Marchand and by Aschoff and Apfelstedt, and described by Eden. In these cases they are undoubtedly fat granules, the preparations being fixed in Miil1er’s or Flemming’s solutions and stained, with osmic acid. My specimen was fixed with formalin, and passed through alcohol and xylol into paraffin, so that it is unlikely that any fat remained. Moreover the granules appear in all the sections stained with haematoxylin and eosin. One slide I re-stained with osmic acid, but the granules remained exactly as before, and did not take on the osmic acid. It may be concluded therefore that they are not fat, however closely they resemble it in appearance and distribution. For glycogen none of my slides were stained, and that also may be put out of account.

Peters describes the presence of red and white blood corpuscles, both normal and degenerated, in the syncytium. But in his case this was an occasional occurrence, although it led him to the rather extreme conclusion that the blood cells were capable of taking part in the formation of syncytium out of their own elements. The appearances in my specimen are, however, quite apart from this. The granules are universally .present throughout the syncytium and symplasma, and in practically no other tissues. Here and there I have seen isolated granules in the decidua or trophoderm cells. In the syncytium, on the other hand, they are so constant as to form in this particular specimen a fairly accurate means of recognizing that tissue.

This uniformity and peculiarity of distribution make it quite clear that it is not an accidental deposit of pigment, and unlikely that it is a degeneration product. The close similarity of their 27 0 Journal of Obstetrics and Gynaecology

arrangement to the fat figured by Marchand and Aschoff, and their translu-cent refractile appearance led me to think that they might be lipoids in a crystalline state. Professor Lorrain Smith Very kindly examined several sections with me, and by means of the polarizing

microscope demonstrated that this idea was wrong. He, however, suggested another explanation which is very probably correct. In

his view these granules, and the very similar granules seen in the spleen, are composed of molecules of blood-pigment adherent to a

A minute droplet of fat or lipoid. If the ti.ssue is fixed and stained for

fat, the granules appear as fat droplets. But in a specimen fixed like mine the lipoid nucleus is dissolved out, and the molecules of pigment are left adhering to one another and forming the granules that take on the eosin. This view I have not been able to test experimentally, but it is certainly the most likely explanation of this interesting appearance. It receives support also from the strong similarity of the granules to those described by jenkinson as occurring in the trophoblast of the sheep. These he proved to consist of haemoglobin derivatives from the ingestion and fragmentation of red blood cells. Bonnet regarded the pigment as haematoidin, but jenkinson failed to confirm that opinion by means of the spectroscope.

This absorption of blood—pigment may, according to Lochhead, be due either to the ingestion of red cells by the trophoblast. or to laking of the blood on the surface and the absorption of the haemoglobin. He is unable to suggest the exact significance of the appearance of the granules. “ Iron—free pigment is probably :1 waste product, while iron—containing pigment is stored in the foetal organs. VVhether the foetus subsequently synthesizes part of the organic iron compound into haemoglobin, or absorbs minute quantities of haemoglobin as such, according to its requirements, is unknown.” At this very early stage the iron requirements of the embryo are probably slight, and the pigment may therefore be in process of being stored in the chorionic villi against later needs.

If it. is true that these granules originally had a fatty nucleus, it indicates that from a very early stage the syncytium is able to take up fat for the benefit of the embrvo. That it does so in the formed placenta is, of course, a recognized fact. It is interesting to note that Eden mentions that he found fat in very occasional areas in the decidua and in the cytotrophoderm. This exact conformity with the distribution of the granules in my specimen is another argument in favour of the truth of Professor Lorrain Smith’s view.

The cytotrophoderm is very similar in both specimens. The enlargement of the cells when they escape from the tips of the villi and spread out into the cell columns is most noticeable in the second specimen. It has also been recognized in several other ova. Johnstone: The Early Human Ovum 27]

The most interesting feature of the cytotrophoderm is the existence in the cell columns of empty spaces, presumably formed by the vacuolation of several adjacent cells (11). This is clearly visible in the first ovum, and it is possible to demonstrate that they do not communicate with the intervillous space. In the second ovum there are numerous smaller spaces entirely devoid of contents, which exactly resemble other spaces lying quite near but full of blood. The presence of these larger spaces in the one ovum, and the coexistence of spaces in the second, some containing blood, others empty, seem to argue strongly in favour of the view that the ramifications of the intervillous space are formed by the opening out of such spaces in the trophoderm. This view receives strong support from the conditions found in the Bryce-Teacher ovum, and these observers believe that the spaces are first formed by the accumulation in the cell columns of the trophodermi-c secretion. This ultimately finds its way to the surface of the trophoderm and

in so doing opens out the spaces, into which the blood then penetrates. Another very strong argument as to the truth of this View is to

be found in the fact that exactly similar spaces may be seen in the trophoderm in the rat and mouse—some, as in my specimen, quite empty, others filled with maternal blood; and the only conceivable way in which the maternal blood could get into the spaces is by their opening out to the surface. This was pointed out by Professor Robinson some years ago, and I have actually seen several of his specimens which demonstrate the point.

The older View of Peters, that the spaces are formed by the outrush of maternal blood under pressure from the decidual vessels, after these had been opened into by the trophodermic activity, does not meet the observed conditions so satisfactorily.

The Embryonic Anlagcn of the Intm-uterine Omtm.

It is most unfortunate that the rupture of the blastocyst and the consequent dislocation of its contents make it impossible to be certain as to the exact nature or arrangement of the two vesicles in this ovum. Before discussing the possible interpretations it may be well to enun-ciate the particular points that must be borne in mind in forming an opinion.

(1) (a) The first vesicle measures 0'35><0'39><O'21 mm.

(17) It consists of two layers, and at one point the outer layer is definitely continuous with the mesoderm of the blastocyst wall. The vesicle is thus. quite clearly attached to the blastocyst wall.

(0) On one side of this rather triangular vesicle there is a distinct thickening, the cells being present in three or more irregular rows. The appearance at once suggests to the eye an embryonic area. The dimensions of this thickened area are 0'21 x 02 mm. 272 Journal of Obstetrics and Gynaecology

(2) (a) The second vesicle measures U‘26><0‘28><0'l4 mm. (b) In structure it is a replica of the first vesicle. (0) It is substantially attached to the blastocyst wall by the

continuity of its outer layer with the blastocyst mesoderm at almost the farthest possible point from the first vesicle.

(d) At a level of about one—third of the way through this Vesicle the cavity is almost completely filled up witl1 cells, which form a floor, so to speak, dividing it into two parts.

(8) The attachments of the respective vesicles to the blastocyst wall are towards opposite ends of one side of that structure.

(4) The embryonic—area-like thickening is on the side of the first vesicle farthest removed from the second vesicle.

(5) Nowhere are the two vesicles in contact. VV here they are most approximated to each other there is no trace of loose cells which might be regarded as the broken remains of an embryonic area between the vesicles.

Naturally the first endeavour is to translate the contents of the blastocyst as parts of a single normal embryo. in doing so several difficulties are met with. In the first place the two vesicles are entirely detached, and secondly there is no indication of any cells that might originally have formed a bond between them. Thirdly, both vesicles are attached to the blastocyst mesoderm at points considerably separated. Fourthly, both vesicles are identical in structure, and there is no suggestion in their structure as to which might be amnion and which yolk sac. Fifthly, while the whole ovum corresponds very -closely in its general development and in the dimensions of the blastocyst to that of Jung, yet each individual vesicle is about the same size as the whole embryonic anlagen (amnion and yolk sac combined) in ]'ung’s specimen. ]ung’s exact figure is 0'25 mm. measured by the number of sections through which the embryonic structures stretched. Later he points out that the yolk sac only stretches through twelve sections, each about 13 micromillimetres thick, giving for that vesicle a length of about 0'15 mm. The exact length of the two vesicles in the present ovum, as measured by the number of sections of known thickness, is 0'35 and 0'28 mm. respectively.

In regard to the first three difficulties it may be argued that the same applies to the Bryce-Teacher ovum. There the two vesicles are even more widely separated, without any trace of connection between them, and each is attached to the mesoderm. So far as it goes that is perfectly true, but that ovum was so young that one would hardly expect any very definitely formed embryonic area intervening between the two sacs, and the attachments to the meso derm are of the most slender variety. Further, the Bryce-Teacher ovum represents a stage before the splitting of the mesenchyme to

form the extra-embryonic coelom, and therefore the attachments Johnstone: The Early Human Ovum 273

may be partly accidental, clue to the shrinkage and condensation of the mesenchyme in the hardening. The present ovum on the other hand represents a stage after the formation of the extra-embryonic coelom, and in spite of that both vesicles have definite attachments to the mesoderm.

In regard to the fourth difficulty, it must be remembered that in Peters’ and _Tung’s ova there is a definite differentiation of the cells at the base of the amnion,indicative of the formation of the embryonic ectoderm. The only situation in which there is anything similar in my ovum is the thickening on the wall of the first vesicle at the side most remote from the second sac.

One other possible view in regard to the attachment of both vesicles is that one attachment represents an amniotic duct. This is a duct or cord of cells connecting the amniotic cavity with the outer surface of the chorion. Eternod, Marchand, and Beneke, have described what they believe to be indications of it in human ova, and Selenka has described it in the ova of apes. According to Keibel it is very doubtful if it is at all a regular condition in the human ovum. Probably not, for it is but “a phylogenetic memory from dim ancestral times.” "In specimens such as Marchand’s there is a depression on the surface of the chorion at the end of this duct. In the case of the second vesicle in my specimen there is no depression, and nothing to support such a view. In regard to the first vesicle I cannot speak so definitely as the chorion has been ruptured in the immediate neighbourhood of the attachment of the sac. But so far as I can see there is no suggestion of a depression. In any case, even if the first sac was clearly proved to be amnion, there still remain the other difficulties in the way of co-ordinating the two sacs as parts of one single embryo.

Failure to interpret the vesicles satisfactorily as parts of a single embryo compelled one to study them as the possible components of a twin ovum. This interpretation involves three assumptions. The first is that the thickening on the wall of the first vesicle is an embryonic area. The second is that on the other side of it there was another sac—-probably the yolk sac——which has been broken up and dispersed. In Fig. 18 there is shown a strand of cells stretching out towards the second vesicle, which, if curled round the opposite way, might be imagined to have formed a closed sac superficial to the thickened area.

The third assumption is that the cells, filling up the cavity of the second vesicle through the thickness of several sections, do really form a floor, and represent the embryonic area of the second embryo. In this event the area must have been cut on the flat, and that part of the vesicle superficial to it would represent the yolk sac, while the part beneath the embryonic area would represent the 274 Journal of Obstetrics and Gynaecology

amnion. This interpretation would bring the second embryo into very close correspondence with the dimensions of jiung’s embryo.

In connection with this view the investigation of the specimen gave a useful example of the value of studying such a specimen, not only by serial sections, but also when possible by means of a model reconstruction (31). For the construction of a wax model on a scale of 1 :.200 of the two vesicles and the side of the blastocyst to which they are attached, I am greatly indebted to Professor Robinson, who- personally made nearly two hundred drawings with the camera lucida, and reconstructed the model. One of the special points that the model cleared up was in connection with a fold or pocket of the first vesicle at the base of the thickened area. The study of some isolated sections, in whi-ch this fold appears as a separate vesicle on the surface of the thickened area, would have led us to regard the fold unhesitatingly as the yolk sac of a first embryo, and would have rendered the interpretation of the specimen as one of uniovular twins perfectly clear and free from any objection. By studying serial sections, however, and still more by the construction of a model in three dimensions it became obvious that this pocket or fold was directly continuous with the rest of the vesicle. VVe were thus led to avoid a somewhat tempting error! In other respects the model fully confirmed our previously formed views as to the relations and shapes of the vesicles and the fact that they do not touch at any point.

The conclusion of the whole question is, I believe, that it is impossible to determine with any approach to certainty in favour of either the one interpretation or the other. The specimen may be one of a single embryo, with or without an amniotic duct, whose component parts have been seriously dislocated by mechanical influences, probably during the removal of the uterus from the body. On the other hand it may be a specimen of uniovular twins at a very early stage of development, the yolk sac(?) of the first embryo being broken up and dispersed. The arguments on either side seem to be fairly well balanced. The decision of any individual observer must remain merely an opinion.

The Age of the two Out.

In neither case have I been able to get any history that would help in determining the age of the ovum. In the first case the husband did not know the date of the last period, although he was quite sure that no period had been missed. Before I could interrogate him more closely he had left the city. In the case of the second specimen, the patient had forgotten the exact dates by the time I came to investigate the question in detail. All she knew was

that the last period had been “ about three weeks before the operaion. FIG. 30.—Ovary and tube with infundibular pregnancy.

FIG. 31.—~Wax reconstruction model (1-100) of the embryonic vesicles and their attachment to the blastocyst wall. The vesicles do not touch at any point. Johnstone: The Early Human Ovum 275

Any estimation of the age must therefore be made by the rather tentative method of comparison with other specimens. The first specimen corresponds closely to ]ung’s in its dimensions, and in the development of the villi. That ovum, according to the table of Bryce and Teacher, whose reckoning I regard as more accurate than that of His in spite of such cases as Giacomini’s “eleven days’ ” ovum, was probably about 14% to 15% days old. I would therefore place the age of my first Specimen at about 15 days.

I have figured and described the corpus luteum, for, as Mall points out, no record of an ovum is complete without it, Tf it is available. In size and development it corresponds fairly closely to the description of the corpus luteum in the second week given by Leopold and Ravano. So far as it goes, therefore, the corpus luteum supports this estimate of the age of the ovum, but I do not think that we are as yet in possession of anything like an accurate standard of the development of that strange organ, and I do not place much importance upon it.

The blastocyst of the second specimen measures 2‘73><6><5'6 mm. The embryonic area proper measures 1'6 mm. The blastecyst therefore corresponds to the specimens of Rossi Doria and Eternod, which Bryce and Teacher place at 18 to 19 days old. The embryo in regard to development lies between “ Gle ” of von Spee and “Klb” of Kroerner and Pfannenstiel. The former is regarded by «Bryce and Teacher as 19 to 20 days old. I think therefore that the second ovum was probably about twenty days old.

In conclusion it is a great pleasure to express my thanks to Professor Harvey Littlejohn and Dr. VVil1iam Fordyce from whom the respective specimens were obtained, and to Dr. Barbour for permission to investigate the first ovum. Professor Arthur Robinson has allowed me to draw freely on his experience and knowledge, and, as stated, personally gave much time to the reconstruction of the model. For all this help, so willingly given, I offer him my heartiest thanks. I am indebted to the Earl of Moray Endowment of the University for a contribution to the expenses of the research.

REFERENCES.

Apfelstedt and Aschofi. Arch. f. Gym, lx, 1896.

Beneke. Momzts. /. Geb. 14. Gym, xix, 1904.

Berkeley and Bonney. Iozmz. of Obsttzt. and G3/11.. of the Brit. Empire, vii, 1905.

Bonnet. Mmmts. f. Geb. u. Gy-n., Xviii, 1903.

Bryce and Teacher. “ Early Development and Embedding of the Human Ovuin.” Glasgow: Maclehose, 1908.

Cova. Arch. f. Gym, lxviii, 1907. 276 Journal of Obstetrics and Gynaecology

Eden. Proc. Roy. Soc. London, IX, 1896.

Fetzer. Arzat-. Anze-"iger, Xxxvii, I910.

Frassi. Arch. f. Micro. Anat, lxx, 1910; lxxi, 1908.

Grosser. Keibel and Mall’s “ Manual of Human Embryology,” London, Lippincott, 1910.

Heine and Hofbauer. Zeitschr. f. Geb. 1;. Gym, lxviii, 1911.

Herzog. Amer. ]0m/n. of Anat., ix, I909.

Heukelom, S. van. Arch. f. Anat. u. Phys, Anat. Abth., I898.

Hofbauer. “Biologie der Plazenta.” VVien, Braiiirmllcr, 1905.

Jung. “ Ei-einbettung beiin inenschlichen Weibe.” Berlin, Karger, 1908.

Keibcl. “Normentaieln zur Entwickelungsgeschichte der V\7irbelthiere.” Pt. 8. Jena, 1908.

Keibel and Ma1l’s “ Manual of Human Enibryology.” London, Lippincott, 1910.

Kroemer. Arch. 1‘. ($3271., lxviii, 1903.

Leopold. “Ein sehr junges nienschlichen Ei.” Leipzig, Hirzel, 1906.

Leopold and Ravano. Arch. f. Gym, lxxxiii, 1907.

Linzenrneier. Arch. f. Gym, cii, Hit. 1, 1914.

Lochhead. Marsl1all’s “Physiology of Reproduction.” London, Longmans, 1910.

Mall, 7)ide Keibel and Mall.

Marchand. Ame‘. Hegfte, xxi, 1903.

Meyer. Zeitschr. f. Gab. 14,. Gym, lxxiv, Hit. 1, I914.

Peters. “Ueber die Einbettung cles menschlichen Eies. 1899.

Robinson. A. Hunterian Lecture. joum. of Amt. and Phj/5., Xxxviii, 1904.

Rossi Doria. Arch. f. Gym, lxxvi, 1905.

Stolper. Morzats. f. Geb. u. Gym, xxiv, 1906.

VVil1iams, J. \Vl1itridge. “Obstetrics,” 1913, edition. London, Appleton.

1)

Leipzig, Deutickc,